---------------------------------------------------------------------------------------------------------------

. * Table 1: Main results

. 
. 
. 
. * Open raw voting and registration data for Table 1

. 
. import delimited "ky_demos.csv", clear
(16 vars, 3,220 obs)

. 
. 
. 
. * Label variables (for reference: they will be reprocessed)

. 
. label var year "Year of election"

. 
. label var code "County, numbered alphabetically per Kentucky Secretary of State"

. 
. label var agegroup "Age: 1 = Under 25; 2 = 25-34; 3 = 35-49; 4 = 50-61; 5 = 62 & Over"

. 
. label var gender "Sex: f = Female; M = male"

. 
. label var totalreg "Total number of registrants in age-gender-county-year"

. 
. label var totalvot "Total number of voters in age-gender-county-year"

. 
. label var demreg "Number of registered Democrats in age-gender-county-year"

. 
. label var demvot "Number of Democratic-registered voters in age-gender-county-year"

. 
. label var repreg "Number of registered Republicans in age-gender-county-year"

. 
. label var repvot "Number of Republican-registered voters in age-gender-county-year"

. 
. label var otherreg "Number of other registrants in age-gender-county-year"

. 
. label var othervot "Number of non-major-registered voters in age-gender-county-year"

. 
. label var timezone "Time Zone: 0 = Eastern Time; 1 = Central Time"

. 
. label var congdist "Congressional District: 1 = 1st; 2 = 2nd; 5 = 5th, 6 = 6th"

. 
. label var expansion "Flag for counties near but not bordering time-zone border"

. 
. 
. 
. 
. 
. * Aggregate data by gender and party groups

. 
. collapse timezone congdist expansion (sum) totalreg totalvot, by(year code agegroup)

. 
. 
. 
. * Create variable marking age-county to panel-correct standard errors

. 
. gen panelvar = agegroup + 10*code

. 
. 
. 
. * Create variable marking county-year for clustering of standard errors

. 
. gen countyyear = year + 10000*code

. 
. 
. 
. * Set up panel by county-age group (cross-section) and year (time-series)

. 
. xtset panelvar year
       panel variable:  panelvar (strongly balanced)
        time variable:  year, 1998 to 2016, but with gaps
                delta:  1 unit

. 
. 
. 
. * Generate turnout (as percentage of registered voters) from raw data

. 
. gen totalpct = 100*totalvot/totalreg

. 
. 
. 
. * Generate turnout (as proportion of registered voters) for GLM

. 
. gen totalprop = totalpct/100

. 
. 
. 
. * Create independent variable of interest, interacting age with time zone

. 
. gen interact = agegroup*timezone

. 
. 
. 
. * Label variables

. 
. label var totalreg "Total number of registrants in age-county-year"

. 
. label var totalvot "Total number of voters in age-county-year"

. 
. label var timezone "Time Zone: 0 = Eastern Time; 1 = Central Time"

. 
. label var congdist "Congressional District: 1 = 1st; 2 = 2nd; 5 = 5th"

. 
. label var panelvar "Code for age group (last digit) and county (earlier digits)"

. 
. label var countyyear "Code for year (last four digits) and county (earlier digits)"

. 
. label var totalpct "Percent of age-county registered voting in election year"

. 
. label var totalprop "Proportion of age-county registered voting in election year"

. 
. label var interact "Interaction of age group and time zone (agegroup*timezone)"

. 
. label var expansion "Flag for counties near but not bordering time-zone border"

. 
. 
. 
. * Estimate model for Table 1's Column I 

. 
. xtpcse totalpct timezone agegroup interact congdist##year if expansion == 1

Number of gaps in sample:  280

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =        980
Time variable:    year                          Number of groups  =         70
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =         14
                                                              avg =         14
                                                              max =         14
Estimated covariances      =      2485          R-squared         =     0.8918
Estimated autocorrelations =         0          Wald chi2(17)     =     916.58
Estimated coefficients     =        45          Prob > chi2       =     0.0000

-------------------------------------------------------------------------------
              |           Panel-corrected
     totalpct |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |   4.403555   .9665486     4.56   0.000     2.509154    6.297955
     agegroup |   8.261356   .4940172    16.72   0.000       7.2931    9.229612
     interact |  -1.443447   .0921592   -15.66   0.000    -1.624075   -1.262818
              |
     congdist |
           2  |  -9.354424   1.855911    -5.04   0.000    -12.99194   -5.716905
           5  |  -7.426155   1.539695    -4.82   0.000     -10.4439   -4.408408
              |
         year |
        1999  |   -44.9022   1.953993   -22.98   0.000    -48.73195   -41.07244
        2000  |   -2.84917   1.953993    -1.46   0.145    -6.678926    .9805849
        2003  |  -22.42809   1.451575   -15.45   0.000    -25.27313   -19.58306
        2004  |   .2770871   1.451575     0.19   0.849    -2.567947    3.122121
        2006  |   -2.33598   1.451575    -1.61   0.108    -5.181014    .5090535
        2007  |  -25.32646   1.451575   -17.45   0.000    -28.17149   -22.48143
        2008  |  -.6319087   1.451575    -0.44   0.663    -3.476943    2.213125
        2010  |  -6.799833   1.451575    -4.68   0.000    -9.644867   -3.954799
        2011  |  -30.94224   1.451575   -21.32   0.000    -33.78727   -28.09721
        2012  |  -1.090197    1.61429    -0.68   0.499    -4.254148    2.073754
        2014  |  -8.025785    1.61429    -4.97   0.000    -11.18974   -4.861834
        2015  |  -28.55719    1.61429   -17.69   0.000    -31.72114   -25.39324
        2016  |   .7457886   1.953993     0.38   0.703    -3.083967    4.575544
              |
congdist#year |
      2 1999  |   19.56359   2.605324     7.51   0.000     14.45725    24.66993
      2 2000  |    8.09573   2.605324     3.11   0.002      2.98939    13.20207
      2 2003  |   9.995849   1.960142     5.10   0.000     6.154042    13.83766
      2 2004  |   11.73268   1.960142     5.99   0.000     7.890876    15.57449
      2 2006  |   4.110764   1.960142     2.10   0.036     .2689563    7.952572
      2 2007  |   10.76805   1.960142     5.49   0.000     6.926238    14.60985
      2 2008  |   11.70785   1.960142     5.97   0.000     7.866037    15.54965
      2 2010  |   6.173169   1.960142     3.15   0.002     2.331361    10.01498
      2 2011  |    8.49807   1.960142     4.34   0.000     4.656262    12.33988
      2 2012  |   5.318337   2.325237     2.29   0.022     .7609563    9.875717
      2 2014  |   2.307081   2.325237     0.99   0.321      -2.2503    6.864461
      2 2015  |   7.938581   2.325237     3.41   0.001     3.381201    12.49596
      2 2016  |   7.774795   2.605324     2.98   0.003     2.668454    12.88113
      5 1999  |   13.23129   1.953993     6.77   0.000     9.401539    17.06105
      5 2000  |   1.944688   1.953993     1.00   0.320    -1.885067    5.774443
      5 2003  |   7.012157   1.451575     4.83   0.000     4.167123    9.857191
      5 2004  |   6.527923   1.451575     4.50   0.000     3.682889    9.372957
      5 2006  |  -.0484184   1.451575    -0.03   0.973    -2.893452    2.796615
      5 2007  |   7.735901   1.451575     5.33   0.000     4.890867    10.58094
      5 2008  |   5.658883   1.451575     3.90   0.000     2.813849    8.503917
      5 2010  |   4.924022   1.451575     3.39   0.001     2.078988    7.769055
      5 2011  |   4.672377   1.451575     3.22   0.001     1.827343    7.517411
      5 2012  |   3.316217    1.61429     2.05   0.040     .1522661    6.480168
      5 2014  |   1.852267    1.61429     1.15   0.251    -1.311684    5.016218
      5 2015  |   4.318936    1.61429     2.68   0.007     1.154985    7.482887
      5 2016  |    3.98396   1.953993     2.04   0.041     .1542047    7.813715
              |
        _cons |   32.87034   2.338697    14.05   0.000     28.28658     37.4541
-------------------------------------------------------------------------------

. 
. 
. 
. * Estimate model for Table 1's Column II

. 
. xtpcse totalpct timezone##agegroup congdist##year if expansion == 1

Number of gaps in sample:  280

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =        980
Time variable:    year                          Number of groups  =         70
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =         14
                                                              avg =         14
                                                              max =         14
Estimated covariances      =      2485          R-squared         =     0.9135
Estimated autocorrelations =         0          Wald chi2(23)     =    4782.84
Estimated coefficients     =        51          Prob > chi2       =     0.0000

-----------------------------------------------------------------------------------
                  |           Panel-corrected
         totalpct |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
       1.timezone |   2.638683   .9027774     2.92   0.003     .8692717    4.408094
                  |
         agegroup |
               2  |   5.094901   .8495895     6.00   0.000     3.429736    6.760066
               3  |   17.49638   1.264217    13.84   0.000     15.01855     19.9742
               4  |   28.23645   1.648769    17.13   0.000     25.00492    31.46798
               5  |   29.73601   2.078809    14.30   0.000     25.66161     33.8104
                  |
timezone#agegroup |
             1 2  |  -.9870284   .1799326    -5.49   0.000     -1.33969   -.6343669
             1 3  |  -2.127685   .3334915    -6.38   0.000    -2.781317   -1.474054
             1 4  |  -4.003759   .3531434   -11.34   0.000    -4.695907   -3.311611
             1 5  |  -5.708868   .3296341   -17.32   0.000    -6.354939   -5.062798
                  |
         congdist |
               2  |  -9.354424   1.855911    -5.04   0.000    -12.99194   -5.716905
               5  |  -7.426155   1.539695    -4.82   0.000     -10.4439   -4.408408
                  |
             year |
            1999  |   -44.9022   1.953993   -22.98   0.000    -48.73195   -41.07244
            2000  |   -2.84917   1.953993    -1.46   0.145    -6.678926    .9805849
            2003  |  -22.42809   1.451575   -15.45   0.000    -25.27313   -19.58306
            2004  |   .2770871   1.451575     0.19   0.849    -2.567947    3.122121
            2006  |   -2.33598   1.451575    -1.61   0.108    -5.181014    .5090535
            2007  |  -25.32646   1.451575   -17.45   0.000    -28.17149   -22.48143
            2008  |  -.6319087   1.451575    -0.44   0.663    -3.476943    2.213125
            2010  |  -6.799833   1.451575    -4.68   0.000    -9.644867   -3.954799
            2011  |  -30.94224   1.451575   -21.32   0.000    -33.78727   -28.09721
            2012  |  -1.090197    1.61429    -0.68   0.499    -4.254148    2.073754
            2014  |  -8.025785    1.61429    -4.97   0.000    -11.18974   -4.861834
            2015  |  -28.55719    1.61429   -17.69   0.000    -31.72114   -25.39324
            2016  |   .7457886   1.953993     0.38   0.703    -3.083967    4.575544
                  |
    congdist#year |
          2 1999  |   19.56359   2.605324     7.51   0.000     14.45725    24.66993
          2 2000  |    8.09573   2.605324     3.11   0.002      2.98939    13.20207
          2 2003  |   9.995849   1.960142     5.10   0.000     6.154042    13.83766
          2 2004  |   11.73268   1.960142     5.99   0.000     7.890876    15.57449
          2 2006  |   4.110764   1.960142     2.10   0.036     .2689563    7.952572
          2 2007  |   10.76805   1.960142     5.49   0.000     6.926238    14.60985
          2 2008  |   11.70785   1.960142     5.97   0.000     7.866037    15.54965
          2 2010  |   6.173169   1.960142     3.15   0.002     2.331361    10.01498
          2 2011  |    8.49807   1.960142     4.34   0.000     4.656262    12.33988
          2 2012  |   5.318337   2.325237     2.29   0.022     .7609563    9.875717
          2 2014  |   2.307081   2.325237     0.99   0.321      -2.2503    6.864461
          2 2015  |   7.938581   2.325237     3.41   0.001     3.381201    12.49596
          2 2016  |   7.774795   2.605324     2.98   0.003     2.668454    12.88113
          5 1999  |   13.23129   1.953993     6.77   0.000     9.401539    17.06105
          5 2000  |   1.944688   1.953993     1.00   0.320    -1.885067    5.774443
          5 2003  |   7.012157   1.451575     4.83   0.000     4.167123    9.857191
          5 2004  |   6.527923   1.451575     4.50   0.000     3.682889    9.372957
          5 2006  |  -.0484184   1.451575    -0.03   0.973    -2.893452    2.796615
          5 2007  |   7.735901   1.451575     5.33   0.000     4.890867    10.58094
          5 2008  |   5.658883   1.451575     3.90   0.000     2.813849    8.503917
          5 2010  |   4.924022   1.451575     3.39   0.001     2.078988    7.769055
          5 2011  |   4.672377   1.451575     3.22   0.001     1.827343    7.517411
          5 2012  |   3.316217    1.61429     2.05   0.040     .1522661    6.480168
          5 2014  |   1.852267    1.61429     1.15   0.251    -1.311684    5.016218
          5 2015  |   4.318936    1.61429     2.68   0.007     1.154985    7.482887
          5 2016  |    3.98396   1.953993     2.04   0.041     .1542047    7.813715
                  |
            _cons |   41.54166   2.109761    19.69   0.000      37.4066    45.67672
-----------------------------------------------------------------------------------

. 
. 
. 
. * Estimate marginal effects of age vs. time-zone interaction for Figure 2 (should correspond to the first ten
> , "Column II" rows of fig2data.csv)

. 
. margins timezone#agegroup

Predictive margins                              Number of obs     =        980
Model VCE    : Panel-corrected

Expression   : Fitted values, predict()

-----------------------------------------------------------------------------------
                  |            Delta-method
                  |     Margin   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
timezone#agegroup |
             0 1  |   28.15849   1.433377    19.64   0.000     25.34912    30.96785
             0 2  |   33.25339     .70669    47.06   0.000      31.8683    34.63847
             0 3  |   45.65486   .3506897   130.19   0.000     44.96752     46.3422
             0 4  |   56.39494   .3878761   145.39   0.000     55.63471    57.15516
             0 5  |   57.89449    1.00241    57.76   0.000      55.9298    59.85918
             1 1  |   30.79717   .8687481    35.45   0.000     29.09445    32.49988
             1 2  |   34.90504   .4557696    76.58   0.000     34.01175    35.79833
             1 3  |   46.16586   .6622702    69.71   0.000     44.86783    47.46388
             1 4  |   55.02986   .8430575    65.27   0.000      53.3775    56.68222
             1 5  |   54.82431   1.217107    45.04   0.000     52.43882    57.20979
-----------------------------------------------------------------------------------

. 
. 
. 
. * Estimate model for Table 1's Column III

. 
. glm totalprop timezone agegroup interact congdist##year if expansion == 1, cl(countyyear) family(binomial)
note: totalprop has noninteger values

Iteration 0:   log pseudolikelihood = -409.67489  
Iteration 1:   log pseudolikelihood =  -409.3677  
Iteration 2:   log pseudolikelihood = -409.36668  
Iteration 3:   log pseudolikelihood = -409.36668  

Generalized linear models                         No. of obs      =        980
Optimization     : ML                             Residual df     =        935
                                                  Scale parameter =          1
Deviance         =  15.81478558                   (1/df) Deviance =   .0169142
Pearson          =  15.77762803                   (1/df) Pearson  =   .0168745

Variance function: V(u) = u*(1-u/1)               [Binomial]
Link function    : g(u) = ln(u/(1-u))             [Logit]

                                                  AIC             =   .9272789
Log pseudolikelihood = -409.3666825               BIC             =  -6424.047

                            (Std. Err. adjusted for 196 clusters in countyyear)
-------------------------------------------------------------------------------
              |               Robust
    totalprop |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |   .2062377   .0591277     3.49   0.000     .0903495    .3221259
     agegroup |   .3716213   .0127109    29.24   0.000     .3467085    .3965342
     interact |  -.0663135   .0162712    -4.08   0.000    -.0982045   -.0344225
              |
     congdist |
           2  |  -.3925205   .1536442    -2.55   0.011    -.6936576   -.0913835
           5  |  -.3088611   .1592725    -1.94   0.052    -.6210295    .0033073
              |
         year |
        1999  |  -2.310756   .2608133    -8.86   0.000    -2.821941   -1.799572
        2000  |  -.1211158   .1299339    -0.93   0.351    -.3757816      .13355
        2003  |  -.9612477   .1491049    -6.45   0.000    -1.253488   -.6690076
        2004  |   .0146278   .1260242     0.12   0.908    -.2323752    .2616307
        2006  |  -.0972277   .2103258    -0.46   0.644    -.5094587    .3150033
        2007  |  -1.096754   .1553945    -7.06   0.000    -1.401321    -.792186
        2008  |   -.024458   .1263812    -0.19   0.847    -.2721606    .2232445
        2010  |   -.285851   .1685241    -1.70   0.090    -.6161521    .0444501
        2011  |  -1.378541   .1371056   -10.05   0.000    -1.647263   -1.109819
        2012  |   -.042602   .1269254    -0.34   0.737    -.2913713    .2061672
        2014  |  -.3368497   .1483669    -2.27   0.023    -.6276435   -.0460559
        2015  |  -1.257377    .139051    -9.04   0.000    -1.529912   -.9848416
        2016  |   .0320164   .1275789     0.25   0.802    -.2180337    .2820665
              |
congdist#year |
      2 1999  |   1.109836   .2947953     3.76   0.000      .532048    1.687624
      2 2000  |   .3435197   .1611669     2.13   0.033     .0276385     .659401
      2 2003  |   .4197125   .1760838     2.38   0.017     .0745946    .7648304
      2 2004  |   .4991916    .160701     3.11   0.002     .1842234    .8141597
      2 2006  |   .1721558   .2388122     0.72   0.471    -.2959075     .640219
      2 2007  |   .4566931    .181261     2.52   0.012     .1014281    .8119582
      2 2008  |   .4972691   .1656495     3.00   0.003      .172602    .8219362
      2 2010  |   .2591584   .2082383     1.24   0.213    -.1489812     .667298
      2 2011  |     .34218   .1684421     2.03   0.042     .0120395    .6723205
      2 2012  |   .2209117   .1602904     1.38   0.168    -.0932517    .5350751
      2 2014  |   .0928823   .1780537     0.52   0.602    -.2560966    .4418611
      2 2015  |   .3193791   .1707233     1.87   0.061    -.0152325    .6539906
      2 2016  |   .3304953    .157463     2.10   0.036     .0218735    .6391171
      5 1999  |   .7365924   .3255701     2.26   0.024     .0984867    1.374698
      5 2000  |     .08246   .2180863     0.38   0.705    -.3449813    .5099012
      5 2003  |   .2827117   .2118274     1.33   0.182    -.1324624    .6978858
      5 2004  |   .2780627   .1855437     1.50   0.134    -.0855964    .6417217
      5 2006  |  -.0047258    .251006    -0.02   0.985    -.4966885    .4872369
      5 2007  |   .3149857   .1821523     1.73   0.084    -.0420263    .6719977
      5 2008  |   .2400677   .2058519     1.17   0.244    -.1633946    .6435301
      5 2010  |   .2056621    .256347     0.80   0.422    -.2967688    .7080929
      5 2011  |    .142943   .1689786     0.85   0.398    -.1882491     .474135
      5 2012  |   .1378121   .1927719     0.71   0.475    -.2400138    .5156381
      5 2014  |   .0718911   .1794834     0.40   0.689    -.2798898    .4236721
      5 2015  |   .1357309   .2007693     0.68   0.499    -.2577698    .5292315
      5 2016  |   .1707661     .18024     0.95   0.343    -.1824977      .52403
              |
        _cons |  -.7962483   .1338449    -5.95   0.000    -1.058579   -.5339172
-------------------------------------------------------------------------------

. 
. 
. 
. * Estimate model for Table 1's Column IV

. 
. glm totalprop timezone##agegroup congdist##year if expansion == 1, cl(countyyear) family(binomial)
note: totalprop has noninteger values

Iteration 0:   log pseudolikelihood = -408.19676  
Iteration 1:   log pseudolikelihood = -407.87498  
Iteration 2:   log pseudolikelihood = -407.87393  
Iteration 3:   log pseudolikelihood = -407.87392  

Generalized linear models                         No. of obs      =        980
Optimization     : ML                             Residual df     =        929
                                                  Scale parameter =          1
Deviance         =  12.82927059                   (1/df) Deviance =   .0138098
Pearson          =  12.73315757                   (1/df) Pearson  =   .0137063

Variance function: V(u) = u*(1-u/1)               [Binomial]
Link function    : g(u) = ln(u/(1-u))             [Logit]

                                                  AIC             =   .9364774
Log pseudolikelihood =  -407.873925               BIC             =  -6385.707

                                (Std. Err. adjusted for 196 clusters in countyyear)
-----------------------------------------------------------------------------------
                  |               Robust
        totalprop |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
       1.timezone |   .1325534   .0498743     2.66   0.008     .0348017    .2303052
                  |
         agegroup |
               2  |   .2552746     .02282    11.19   0.000     .2105482    .3000011
               3  |   .8154553   .0327686    24.89   0.000     .7512299    .8796807
               4  |   1.280641   .0412243    31.07   0.000     1.199843    1.361439
               5  |   1.346682   .0537424    25.06   0.000     1.241348    1.452015
                  |
timezone#agegroup |
             1 2  |  -.0572404    .031337    -1.83   0.068    -.1186597    .0041789
             1 3  |  -.1120308   .0429836    -2.61   0.009    -.1962771   -.0277845
             1 4  |  -.1895397   .0532594    -3.56   0.000    -.2939263   -.0851531
             1 5  |   -.264669   .0699668    -3.78   0.000    -.4018014   -.1275366
                  |
         congdist |
               2  |  -.3942582   .1543075    -2.56   0.011    -.6966954    -.091821
               5  |  -.3103236   .1599684    -1.94   0.052     -.623856    .0032088
                  |
             year |
            1999  |  -2.314928   .2610268    -8.87   0.000    -2.826531   -1.803325
            2000  |  -.1216563   .1305189    -0.93   0.351    -.3774686    .1341559
            2003  |   -.964625   .1496616    -6.45   0.000    -1.257956   -.6712936
            2004  |   .0146857   .1265946     0.12   0.908    -.2334351    .2628065
            2006  |  -.0976821   .2112848    -0.46   0.644    -.5117926    .3164285
            2007  |  -1.100373   .1559065    -7.06   0.000    -1.405944    -.794802
            2008  |  -.0245811   .1269537    -0.19   0.846    -.2734057    .2242436
            2010  |    -.28712   .1692678    -1.70   0.090    -.6188788    .0446388
            2011  |  -1.382482   .1376052   -10.05   0.000    -1.652183   -1.112781
            2012  |  -.0428169   .1275004    -0.34   0.737     -.292713    .2070793
            2014  |  -.3383459   .1490255    -2.27   0.023    -.6304306   -.0462613
            2015  |   -1.26118   .1395738    -9.04   0.000     -1.53474   -.9876204
            2016  |   .0321623   .1281587     0.25   0.802    -.2190242    .2833487
                  |
    congdist#year |
          2 1999  |   1.111855   .2950919     3.77   0.000     .5334855    1.690225
          2 2000  |   .3449901   .1618706     2.13   0.033     .0277295    .6622507
          2 2003  |    .421498   .1767408     2.38   0.017     .0750925    .7679035
          2 2004  |   .5014339   .1614069     3.11   0.002     .1850823    .8177856
          2 2006  |   .1729161   .2398761     0.72   0.471    -.2972324    .6430645
          2 2007  |   .4585437   .1818861     2.52   0.012     .1020536    .8150339
          2 2008  |   .4994937   .1663826     3.00   0.003     .1733897    .8255977
          2 2010  |   .2603332   .2091173     1.24   0.213    -.1495292    .6701957
          2 2011  |   .3439683   .1690512     2.03   0.042      .012634    .6753027
          2 2012  |   .2218762   .1609824     1.38   0.168    -.0936434    .5373959
          2 2014  |   .0935473   .1788113     0.52   0.601    -.2569165     .444011
          2 2015  |   .3210621   .1713536     1.87   0.061    -.0147849    .6569091
          2 2016  |   .3319227   .1581567     2.10   0.036     .0219413     .641904
          5 1999  |   .7388007    .325762     2.27   0.023     .1003189    1.377283
          5 2000  |   .0828408    .219012     0.38   0.705    -.3464148    .5120964
          5 2003  |   .2841197   .2124933     1.34   0.181    -.1323595    .7005989
          5 2004  |   .2793471   .1863911     1.50   0.134    -.0859728     .644667
          5 2006  |  -.0046819   .2521089    -0.02   0.985    -.4988063    .4894425
          5 2007  |   .3165033   .1827975     1.73   0.083    -.0417732    .6747798
          5 2008  |   .2411602   .2067916     1.17   0.244     -.164144    .6464644
          5 2010  |   .2066053   .2574175     0.80   0.422    -.2979236    .7111343
          5 2011  |   .1446575    .169615     0.85   0.394    -.1877818    .4770968
          5 2012  |   .1384398   .1936254     0.71   0.475    -.2410589    .5179386
          5 2014  |   .0723982   .1802605     0.40   0.688    -.2809059    .4257024
          5 2015  |   .1372773   .2013221     0.68   0.495    -.2573067    .5318613
          5 2016  |   .1715289   .1810511     0.95   0.343    -.1833248    .5263825
                  |
            _cons |  -.4217185   .1321662    -3.19   0.001    -.6807595   -.1626776
-----------------------------------------------------------------------------------

. 
. 
. 
. * Estimate marginal effects of age vs. time-zone interaction for Figure 2 (should correspond to the last ten,
>  "Column IV" rows of fig2data.csv)

. 
. margins timezone#agegroup

Predictive margins                              Number of obs     =        980
Model VCE    : Robust

Expression   : Predicted mean totalprop, predict()

-----------------------------------------------------------------------------------
                  |            Delta-method
                  |     Margin   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
timezone#agegroup |
             0 1  |   .2817243   .0070713    39.84   0.000     .2678648    .2955838
             0 2  |   .3328465   .0047752    69.70   0.000     .3234874    .3422056
             0 3  |   .4567547   .0046656    97.90   0.000     .4476102    .4658991
             0 4  |    .563544   .0045331   124.32   0.000     .5546593    .5724287
             0 5  |   .5784223   .0068036    85.02   0.000     .5650875    .5917572
             1 1  |   .3077274   .0063366    48.56   0.000     .2953078    .3201469
             1 2  |   .3487109   .0051384    67.86   0.000     .3386399     .358782
             1 3  |   .4614701   .0051835    89.03   0.000     .4513106    .4716297
             1 4  |   .5506078    .005079   108.41   0.000     .5406532    .5605625
             1 5  |   .5485375   .0064541    84.99   0.000     .5358876    .5611873
-----------------------------------------------------------------------------------

. 
. 
. 
. 
. 
. 
. 
. * Appendix: Analysis for extreme age groups separately

. 
. 
. 
. xtpcse totalpct timezone congdist##year if agegroup == 1 & expansion == 1

Number of gaps in sample:  56

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =        196
Time variable:    year                          Number of groups  =         14
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =         14
                                                              avg =         14
                                                              max =         14
Estimated covariances      =       105          R-squared         =     0.9227
Estimated autocorrelations =         0          Wald chi2(15)     =     224.87
Estimated coefficients     =        43          Prob > chi2       =     0.0000

-------------------------------------------------------------------------------
              |           Panel-corrected
     totalpct |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |   2.815518   .9243441     3.05   0.002     1.003837    4.627199
              |
     congdist |
           2  |  -14.17393   1.813369    -7.82   0.000    -17.72807   -10.61979
           5  |  -5.655465   1.543539    -3.66   0.000    -8.680746   -2.630183
              |
         year |
        1999  |  -38.38944   1.910538   -20.09   0.000    -42.13402   -34.64485
        2000  |   .3886642   1.910538     0.20   0.839    -3.355921    4.133249
        2003  |   -23.0872   1.616374   -14.28   0.000    -26.25524   -19.91916
        2004  |   8.303138   1.616374     5.14   0.000     5.135102    11.47117
        2006  |  -5.197975   1.616374    -3.22   0.001    -8.366011   -2.029939
        2007  |  -26.02441   1.616374   -16.10   0.000    -29.19245   -22.85638
        2008  |   5.071018   1.616374     3.14   0.002     1.902982    8.239054
        2010  |  -10.68669   1.616374    -6.61   0.000    -13.85472   -7.518653
        2011  |  -31.55098   1.616374   -19.52   0.000    -34.71901   -28.38294
        2012  |   1.438483   1.755935     0.82   0.413    -2.003087    4.880053
        2014  |  -10.74457   1.755935    -6.12   0.000    -14.18614      -7.303
        2015  |  -27.92324   1.755935   -15.90   0.000    -31.36481   -24.48167
        2016  |   10.61235   1.910538     5.55   0.000     6.867768    14.35694
              |
congdist#year |
      2 1999  |   20.28834   2.547383     7.96   0.000     15.29556    25.28112
      2 2000  |   11.94919   2.547383     4.69   0.000     6.956414    16.94197
      2 2003  |   11.50994   2.240608     5.14   0.000     7.118432    15.90145
      2 2004  |   13.38537   2.240608     5.97   0.000     8.993859    17.77688
      2 2006  |   5.982899   2.240608     2.67   0.008     1.591389    10.37441
      2 2007  |   12.74082   2.240608     5.69   0.000     8.349309    17.13233
      2 2008  |   15.39793   2.240608     6.87   0.000     11.00642    19.78944
      2 2010  |    7.95011   2.240608     3.55   0.000     3.558599    12.34162
      2 2011  |   12.42749   2.240608     5.55   0.000     8.035981      16.819
      2 2012  |   8.350679   2.632625     3.17   0.002     3.190829    13.51053
      2 2014  |    4.37435   2.632625     1.66   0.097    -.7855003      9.5342
      2 2015  |   10.95165   2.632625     4.16   0.000     5.791801     16.1115
      2 2016  |   9.425965   2.547383     3.70   0.000     4.433185    14.41874
      5 1999  |   11.61454   1.910538     6.08   0.000      7.86995    15.35912
      5 2000  |   1.062695   1.910538     0.56   0.578     -2.68189    4.807279
      5 2003  |   5.070998   1.616374     3.14   0.002     1.902963    8.239034
      5 2004  |   4.713407   1.616374     2.92   0.004     1.545371    7.881442
      5 2006  |  -1.424075   1.616374    -0.88   0.378    -4.592111     1.74396
      5 2007  |   7.228666   1.616374     4.47   0.000      4.06063     10.3967
      5 2008  |   3.850477   1.616374     2.38   0.017     .6824416    7.018513
      5 2010  |   4.398811   1.616374     2.72   0.007     1.230775    7.566847
      5 2011  |    5.86747   1.616374     3.63   0.000     2.699434    9.035505
      5 2012  |   2.682874   1.755935     1.53   0.127     -.758696    6.124444
      5 2014  |   .8376858   1.755935     0.48   0.633    -2.603884    4.279256
      5 2015  |   4.018871   1.755935     2.29   0.022     .5773009     7.46044
      5 2016  |   2.434453   1.910538     1.27   0.203    -1.310131    6.179038
              |
        _cons |   41.07262   1.543539    26.61   0.000     38.04734     44.0979
-------------------------------------------------------------------------------

. 
. xtpcse totalpct timezone congdist##year if agegroup == 5 & expansion == 1

Number of gaps in sample:  56

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =        196
Time variable:    year                          Number of groups  =         14
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =         14
                                                              avg =         14
                                                              max =         14
Estimated covariances      =       105          R-squared         =     0.8825
Estimated autocorrelations =         0          Wald chi2(15)     =     211.42
Estimated coefficients     =        43          Prob > chi2       =     0.0000

-------------------------------------------------------------------------------
              |           Panel-corrected
     totalpct |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |  -3.338466   .7525842    -4.44   0.000    -4.813504   -1.863428
              |
     congdist |
           2  |  -6.052956     2.3259    -2.60   0.009    -10.61164   -1.494276
           5  |  -10.16726   1.869292    -5.44   0.000      -13.831   -6.503514
              |
         year |
        1999  |  -47.32007   2.431749   -19.46   0.000    -52.08621   -42.55393
        2000  |  -5.951958   2.431749    -2.45   0.014     -10.7181   -1.185818
        2003  |  -20.72991   1.889551   -10.97   0.000    -24.43336   -17.02645
        2004  |  -6.780721   1.889551    -3.59   0.000    -10.48417   -3.077268
        2006  |  -1.729602   1.889551    -0.92   0.360    -5.433055     1.97385
        2007  |  -23.08699   1.889551   -12.22   0.000    -26.79044   -19.38353
        2008  |  -5.029965   1.889551    -2.66   0.008    -8.733417   -1.326512
        2010  |  -2.675086   1.889551    -1.42   0.157    -6.378539    1.028366
        2011  |  -25.28156   1.889551   -13.38   0.000    -28.98501   -21.57811
        2012  |  -.3572485   1.968984    -0.18   0.856    -4.216387     3.50189
        2014  |  -2.204211   1.968984    -1.12   0.263    -6.063349    1.654928
        2015  |  -23.97352   1.968984   -12.18   0.000    -27.83266   -20.11438
        2016  |    -.95556   2.431749    -0.39   0.694      -5.7217     3.81058
              |
congdist#year |
      2 1999  |   19.75233   3.242332     6.09   0.000     13.39748    26.10719
      2 2000  |   6.812869   3.242332     2.10   0.036      .458016    13.16772
      2 2003  |   11.36833   2.573992     4.42   0.000     6.323402    16.41327
      2 2004  |   11.72224   2.573992     4.55   0.000     6.677311    16.76718
      2 2006  |   5.524857   2.573992     2.15   0.032      .479924    10.56979
      2 2007  |   10.90746   2.573992     4.24   0.000     5.862529    15.95239
      2 2008  |   10.30419   2.573992     4.00   0.000     5.259254    15.34912
      2 2010  |   7.015858   2.573992     2.73   0.006     1.970925    12.06079
      2 2011  |    8.07317   2.573992     3.14   0.002     3.028238     13.1181
      2 2012  |   4.632281   2.805966     1.65   0.099    -.8673105    10.13187
      2 2014  |   3.025957   2.805966     1.08   0.281    -2.473634    8.525548
      2 2015  |   8.742269   2.805966     3.12   0.002     3.242678    14.24186
      2 2016  |   6.407633   3.242332     1.98   0.048     .0527803    12.76249
      5 1999  |   14.53642   2.431749     5.98   0.000     9.770276    19.30256
      5 2000  |   4.007053   2.431749     1.65   0.099    -.7590865    8.773193
      5 2003  |   9.393655   1.889551     4.97   0.000     5.690203    13.09711
      5 2004  |   8.631811   1.889551     4.57   0.000     4.928358    12.33526
      5 2006  |   1.508298   1.889551     0.80   0.425    -2.195154    5.211751
      5 2007  |   9.957776   1.889551     5.27   0.000     6.254324    13.66123
      5 2008  |   9.446438   1.889551     5.00   0.000     5.742986    13.14989
      5 2010  |    7.27823   1.889551     3.85   0.000     3.574777    10.98168
      5 2011  |   5.870185   1.889551     3.11   0.002     2.166733    9.573638
      5 2012  |    5.74803   1.968984     2.92   0.004     1.888892    9.607168
      5 2014  |   3.193339   1.968984     1.62   0.105    -.6657997    7.052477
      5 2015  |   5.440199   1.968984     2.76   0.006     1.581061    9.299338
      5 2016  |   5.346656   2.431749     2.20   0.028     .5805163     10.1128
              |
        _cons |   69.16196   1.869292    37.00   0.000     65.49822    72.82571
-------------------------------------------------------------------------------

. 
. 
. 
. glm totalprop timezone congdist##year if agegroup == 1 & expansion == 1, cl(countyyear) family(binomial)
note: totalprop has noninteger values

Iteration 0:   log pseudolikelihood = -73.412016  
Iteration 1:   log pseudolikelihood = -73.012517  
Iteration 2:   log pseudolikelihood = -73.010514  
Iteration 3:   log pseudolikelihood =  -73.01051  

Generalized linear models                         No. of obs      =        196
Optimization     : ML                             Residual df     =        153
                                                  Scale parameter =          1
Deviance         =  1.798539555                   (1/df) Deviance =   .0117552
Pearson          =  1.816949821                   (1/df) Pearson  =   .0118755

Variance function: V(u) = u*(1-u/1)               [Binomial]
Link function    : g(u) = ln(u/(1-u))             [Logit]

                                                  AIC             =   1.183781
Log pseudolikelihood = -73.01051022               BIC             =   -805.753

                            (Std. Err. adjusted for 196 clusters in countyyear)
-------------------------------------------------------------------------------
              |               Robust
    totalprop |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |   .1523043   .0373911     4.07   0.000      .079019    .2255896
              |
     congdist |
           2  |  -.6079632   .2161448    -2.81   0.005    -1.031599   -.1843271
           5  |   -.202745   .3293494    -0.62   0.538    -.8482581     .442768
              |
         year |
        1999  |  -2.598376   .3174486    -8.19   0.000    -3.220564   -1.976188
        2000  |   .0157675   .1862363     0.08   0.933    -.3492489    .3807839
        2003  |  -1.097758   .1931034    -5.68   0.000    -1.476234   -.7192822
        2004  |   .3433297   .1819153     1.89   0.059    -.0132176    .6998771
        2006  |  -.2068963   .2767907    -0.75   0.455    -.7493961    .3356036
        2007  |  -1.291981   .1901414    -6.79   0.000    -1.664652   -.9193108
        2008  |   .2138586   .1741394     1.23   0.219    -.1274484    .5551656
        2010  |  -.4479974   .2219276    -2.02   0.044    -.8829675   -.0130272
        2011  |  -1.745493   .1896563    -9.20   0.000    -2.117213   -1.373774
        2012  |    .073256    .180043     0.41   0.684    -.2796218    .4261338
        2014  |  -.4473592   .1990078    -2.25   0.025    -.8374073    -.057311
        2015  |  -1.442634   .1795096    -8.04   0.000    -1.794467   -1.090802
        2016  |   .4262113   .1899354     2.24   0.025     .0539448    .7984778
              |
congdist#year |
      2 1999  |   1.342562   .3733231     3.60   0.000     .6108618    2.074262
      2 2000  |   .5367977    .229869     2.34   0.020     .0862628    .9873327
      2 2003  |   .4207047   .2460083     1.71   0.087    -.0614627     .902872
      2 2004  |   .5862868   .2299782     2.55   0.011     .1355378    1.037036
      2 2006  |   .2445516   .3221514     0.76   0.448    -.3868534    .8759567
      2 2007  |   .4869623   .2342834     2.08   0.038     .0277754    .9461493
      2 2008  |   .6668951   .2268443     2.94   0.003     .2222884    1.111502
      2 2010  |   .3079529   .2750986     1.12   0.263    -.2312305    .8471363
      2 2011  |   .3814489   .2400257     1.59   0.112    -.0889927    .8518906
      2 2012  |   .3680372   .2264935     1.62   0.104    -.0758818    .8119563
      2 2014  |   .1070345   .2455378     0.44   0.663    -.3742107    .5882797
      2 2015  |   .3222719   .2348305     1.37   0.170    -.1379875    .7825313
      2 2016  |   .4394751   .2341505     1.88   0.061    -.0194515    .8984016
      5 1999  |   .8410085    .526188     1.60   0.110    -.1903011    1.872318
      5 2000  |   .0471233   .3426083     0.14   0.891    -.6243767    .7186232
      5 2003  |   .1410373   .3398398     0.42   0.678    -.5250364     .807111
      5 2004  |   .1947502   .3373862     0.58   0.564    -.4665147    .8560151
      5 2006  |  -.0977074   .4130215    -0.24   0.813    -.9072148    .7117999
      5 2007  |   .2800335   .3494895     0.80   0.423    -.4049533    .9650203
      5 2008  |   .1594647   .3571253     0.45   0.655    -.5404881    .8594174
      5 2010  |   .1596361   .4361207     0.37   0.714    -.6951448    1.014417
      5 2011  |   .1190694   .3389917     0.35   0.725    -.5453422    .7834811
      5 2012  |    .102765   .3594526     0.29   0.775    -.6017491    .8072792
      5 2014  |  -.0234685   .3558566    -0.07   0.947    -.7209346    .6739976
      5 2015  |   .0039871   .3426106     0.01   0.991    -.6675173    .6754916
      5 2016  |   .1130803    .342502     0.33   0.741    -.5582113     .784372
              |
        _cons |  -.3980074   .1759447    -2.26   0.024    -.7428527   -.0531621
-------------------------------------------------------------------------------

. 
. glm totalprop timezone congdist##year if agegroup == 5 & expansion == 1, cl(countyyear) family(binomial)
note: totalprop has noninteger values

Iteration 0:   log pseudolikelihood = -85.106026  
Iteration 1:   log pseudolikelihood = -85.099975  
Iteration 2:   log pseudolikelihood = -85.099969  

Generalized linear models                         No. of obs      =        196
Optimization     : ML                             Residual df     =        153
                                                  Scale parameter =          1
Deviance         =  1.510991755                   (1/df) Deviance =   .0098758
Pearson          =  1.510757504                   (1/df) Pearson  =   .0098742

Variance function: V(u) = u*(1-u/1)               [Binomial]
Link function    : g(u) = ln(u/(1-u))             [Logit]

                                                  AIC             =   1.307143
Log pseudolikelihood = -85.09996879               BIC             =  -806.0406

                            (Std. Err. adjusted for 196 clusters in countyyear)
-------------------------------------------------------------------------------
              |               Robust
    totalprop |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |  -.1428865   .0324096    -4.41   0.000    -.2064081   -.0793648
              |
     congdist |
           2  |  -.2605674   .1185958    -2.20   0.028     -.493011   -.0281239
           5  |  -.4345784   .0945126    -4.60   0.000    -.6198198   -.2493371
              |
         year |
        1999  |  -2.138036   .2498007    -8.56   0.000    -2.627636   -1.648435
        2000  |   -.255323   .0896301    -2.85   0.004    -.4309948   -.0796513
        2003  |  -.8545268    .124399    -6.87   0.000    -1.098344   -.6107093
        2004  |   -.290403   .0991257    -2.93   0.003    -.4846857   -.0961202
        2006  |  -.0746739   .1889975    -0.40   0.693    -.4451022    .2957545
        2007  |  -.9499706   .1241603    -7.65   0.000     -1.19332   -.7066209
        2008  |  -.2169015   .1034582    -2.10   0.036    -.4196759   -.0141271
        2010  |  -.1159801   .1461255    -0.79   0.427    -.4023808    .1704206
        2011  |   -1.03989   .1185838    -8.77   0.000     -1.27231   -.8074697
        2012  |  -.0125351    .097335    -0.13   0.898    -.2033082    .1782381
        2014  |  -.0948042   .1200807    -0.79   0.430    -.3301582    .1405497
        2015  |  -.9872083    .108709    -9.08   0.000    -1.200274   -.7741424
        2016  |  -.0421992   .0895112    -0.47   0.637    -.2176379    .1332395
              |
congdist#year |
      2 1999  |   1.002242   .2851885     3.51   0.000      .443283    1.561201
      2 2000  |   .2919298     .14159     2.06   0.039     .0144184    .5694412
      2 2003  |   .4714947   .1554278     3.03   0.002     .1668619    .7761275
      2 2004  |   .5052578   .1391294     3.63   0.000     .2325692    .7779465
      2 2006  |   .2385237   .2131332     1.12   0.263    -.1792097    .6562571
      2 2007  |   .4540264   .1533157     2.96   0.003     .1535331    .7545197
      2 2008  |   .4467183   .1459775     3.06   0.002     .1606077    .7328289
      2 2010  |   .3040013   .1836371     1.66   0.098    -.0559208    .6639234
      2 2011  |   .3415041   .1507875     2.26   0.024      .045966    .6370422
      2 2012  |   .1972512   .1469645     1.34   0.180    -.0907939    .4852963
      2 2014  |   .1298316   .1484803     0.87   0.382    -.1611845    .4208476
      2 2015  |   .3693591    .139407     2.65   0.008     .0961264    .6425919
      2 2016  |   .2804877   .1217875     2.30   0.021     .0417886    .5191867
      5 1999  |   .7392611   .2625572     2.82   0.005     .2246584    1.253864
      5 2000  |   .1754598   .1858831     0.94   0.345    -.1888644     .539784
      5 2003  |   .3970498   .2158443     1.84   0.066    -.0259973    .8200968
      5 2004  |   .3674913   .1669205     2.20   0.028     .0403332    .6946493
      5 2006  |   .0655332   .1956598     0.33   0.738    -.3179531    .4490194
      5 2007  |   .4204654   .1400985     3.00   0.003     .1458775    .6950534
      5 2008  |   .4030523   .1586454     2.54   0.011     .0921131    .7139916
      5 2010  |   .3101853    .172372     1.80   0.072    -.0276577    .6480283
      5 2011  |   .2532861     .13667     1.85   0.064    -.0145822    .5211544
      5 2012  |   .2409236   .1591764     1.51   0.130    -.0710563    .5529035
      5 2014  |   .1358492   .1225635     1.11   0.268    -.1043707    .3760692
      5 2015  |   .2371848   .1629637     1.46   0.146    -.0822182    .5565877
      5 2016  |   .2272565   .1250181     1.82   0.069    -.0177746    .4722875
              |
        _cons |   .7983249   .0931929     8.57   0.000     .6156701    .9809797
-------------------------------------------------------------------------------

. 
. 
. 
. 
. 
. 
. 
. * Appendix: Analysis only for (non-)presidential election years

. 
. 
. 
. xtpcse totalpct timezone agegroup interact congdist##year if mod(year, 4) == 0 & expansion == 1

Number of gaps in sample:  280

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =        350
Time variable:    year                          Number of groups  =         70
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =          5
                                                              avg =          5
                                                              max =          5
Estimated covariances      =      2485          R-squared         =     0.7576
Estimated autocorrelations =         0          Wald chi2(8)      =   15460.00
Estimated coefficients     =        18          Prob > chi2       =     0.0000

-------------------------------------------------------------------------------
              |           Panel-corrected
     totalpct |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |    1.84667   .8227856     2.24   0.025     .2340394      3.4593
     agegroup |   6.239882   .4712882    13.24   0.000     5.316174    7.163589
     interact |   -1.21015   .1150805   -10.52   0.000    -1.435704   -.9845967
              |
     congdist |
           2  |  -2.290358   1.234264    -1.86   0.064     -4.70947    .1287542
           5  |  -7.338463   1.153119    -6.36   0.000    -9.598535   -5.078391
              |
         year |
        2004  |   2.662009   .9601888     2.77   0.006     .7800732    4.543944
        2008  |   1.753013   .9601888     1.83   0.068    -.1289226    3.634948
        2012  |   1.016175   1.134443     0.90   0.370    -1.207293    3.239643
        2016  |   3.594959   1.152693     3.12   0.002     1.335722    5.854196
              |
congdist#year |
      2 2004  |   4.204369   1.308667     3.21   0.001     1.639428     6.76931
      2 2008  |    4.17953   1.308667     3.19   0.001     1.614589    6.744471
      2 2012  |  -1.798786   1.691728    -1.06   0.288    -5.114513     1.51694
      2 2016  |  -.3209359   1.536924    -0.21   0.835    -3.333252     2.69138
      5 2004  |   5.047484   .9601888     5.26   0.000     3.165549    6.929419
      5 2008  |   4.178444   .9601888     4.35   0.000     2.296508    6.060379
      5 2012  |   2.114327   1.134443     1.86   0.062    -.1091406    4.337795
      5 2016  |   2.039272   1.152693     1.77   0.077     -.219965    4.298509
              |
        _cons |   37.94259   1.869871    20.29   0.000     34.27771    41.60747
-------------------------------------------------------------------------------

. 
. xtpcse totalpct timezone agegroup interact congdist##year if mod(year, 4) != 0 & expansion == 1

Number of gaps in sample:  280

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =        630
Time variable:    year                          Number of groups  =         70
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =          9
                                                              avg =          9
                                                              max =          9
Estimated covariances      =      2485          R-squared         =     0.8935
Estimated autocorrelations =         0          Wald chi2(12)     =    3885.91
Estimated coefficients     =        30          Prob > chi2       =     0.0000

-------------------------------------------------------------------------------
              |           Panel-corrected
     totalpct |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |   5.786371   1.217274     4.75   0.000     3.400558    8.172185
     agegroup |   9.384397   .3600203    26.07   0.000      8.67877    10.09002
     interact |  -1.573056   .1060096   -14.84   0.000    -1.780831   -1.365281
              |
     congdist |
           2  |  -8.802208   2.109907    -4.17   0.000    -12.93755   -4.666866
           5  |  -6.432167    1.80449    -3.56   0.000    -9.968901   -2.895432
              |
         year |
        1999  |   -44.9022   2.226831   -20.16   0.000     -49.2667   -40.53769
        2003  |   -22.1796   1.667502   -13.30   0.000    -25.44784   -18.91135
        2006  |  -2.087483   1.667502    -1.25   0.211    -5.355728    1.180761
        2007  |  -25.07796   1.667502   -15.04   0.000    -28.34621   -21.80972
        2010  |  -6.551336   1.667502    -3.93   0.000     -9.81958   -3.283091
        2011  |  -30.69374   1.667502   -18.41   0.000    -33.96199    -27.4255
        2014  |   -7.62819   1.855729    -4.11   0.000    -11.26535   -3.991027
        2015  |  -28.15959   1.855729   -15.17   0.000    -31.79675   -24.52243
              |
congdist#year |
      2 1999  |   19.56359   2.969108     6.59   0.000     13.74424    25.38293
      2 2003  |   9.692131    2.25855     4.29   0.000     5.265453    14.11881
      2 2006  |   3.807045    2.25855     1.69   0.092    -.6196321    8.233723
      2 2007  |   10.46433    2.25855     4.63   0.000      6.03765      14.891
      2 2010  |    5.86945    2.25855     2.60   0.009     1.442772    10.29613
      2 2011  |   8.194351    2.25855     3.63   0.000     3.767673    12.62103
      2 2014  |   1.783264   2.686638     0.66   0.507    -3.482449    7.048978
      2 2015  |   7.414765   2.686638     2.76   0.006     2.149051    12.68048
      5 1999  |   13.23129   2.226831     5.94   0.000     8.866786     17.5958
      5 2003  |    6.76366   1.667502     4.06   0.000     3.495415     10.0319
      5 2006  |  -.2969156   1.667502    -0.18   0.859     -3.56516    2.971329
      5 2007  |   7.487404   1.667502     4.49   0.000      4.21916    10.75565
      5 2010  |   4.675524   1.667502     2.80   0.005      1.40728    7.943769
      5 2011  |    4.42388   1.667502     2.65   0.008     1.155635    7.692124
      5 2014  |   1.454671   1.855729     0.78   0.433    -2.182491    5.091834
      5 2015  |   3.921341   1.855729     2.11   0.035     .2841782    7.558504
              |
        _cons |   28.50723    2.25443    12.64   0.000     24.08862    32.92583
-------------------------------------------------------------------------------

. 
. 
. 
. xtpcse totalpct timezone##agegroup congdist##year if mod(year, 4) == 0 & expansion == 1

Number of gaps in sample:  280

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =        350
Time variable:    year                          Number of groups  =         70
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =          5
                                                              avg =          5
                                                              max =          5
Estimated covariances      =      2485          R-squared         =     0.8566
Estimated autocorrelations =         0          Wald chi2(9)      =    1867.62
Estimated coefficients     =        24          Prob > chi2       =     0.0000

-----------------------------------------------------------------------------------
                  |           Panel-corrected
         totalpct |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
       1.timezone |    .505369   .7249071     0.70   0.486    -.9154229    1.926161
                  |
         agegroup |
               2  |   1.778146   1.047988     1.70   0.090    -.2758725    3.832164
               3  |   13.18038   1.330372     9.91   0.000      10.5729    15.78786
               4  |   22.12794    1.59078    13.91   0.000     19.01007    25.24581
               5  |   21.02451   2.010956    10.45   0.000     17.08311    24.96591
                  |
timezone#agegroup |
             1 2  |   -1.18206   .3441042    -3.44   0.001    -1.856492   -.5076279
             1 3  |  -2.067568   .2370524    -8.72   0.000    -2.532182   -1.602954
             1 4  |  -3.108684   .5182506    -6.00   0.000    -4.124437   -2.092932
             1 5  |  -5.087439     .47019   -10.82   0.000    -6.008995   -4.165884
                  |
         congdist |
               2  |  -2.290358   1.234264    -1.86   0.064     -4.70947    .1287542
               5  |  -7.338463   1.153119    -6.36   0.000    -9.598535   -5.078391
                  |
             year |
            2004  |   2.662009   .9601888     2.77   0.006     .7800732    4.543944
            2008  |   1.753013   .9601888     1.83   0.068    -.1289226    3.634948
            2012  |   1.016175   1.134443     0.90   0.370    -1.207293    3.239643
            2016  |   3.594959   1.152693     3.12   0.002     1.335722    5.854196
                  |
    congdist#year |
          2 2004  |   4.204369   1.308667     3.21   0.001     1.639428     6.76931
          2 2008  |    4.17953   1.308667     3.19   0.001     1.614589    6.744471
          2 2012  |  -1.798786   1.691728    -1.06   0.288    -5.114513     1.51694
          2 2016  |  -.3209359   1.536924    -0.21   0.835    -3.333252     2.69138
          5 2004  |   5.047484   .9601888     5.26   0.000     3.165549    6.929419
          5 2008  |   4.178444   .9601888     4.35   0.000     2.296508    6.060379
          5 2012  |   2.114327   1.134443     1.86   0.062    -.1091406    4.337795
          5 2016  |   2.039272   1.152693     1.77   0.077     -.219965    4.298509
                  |
            _cons |   45.04003   1.583839    28.44   0.000     41.93577     48.1443
-----------------------------------------------------------------------------------

. 
. xtpcse totalpct timezone##agegroup congdist##year if mod(year, 4) != 0 & expansion == 1

Number of gaps in sample:  280

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =        630
Time variable:    year                          Number of groups  =         70
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =          9
                                                              avg =          9
                                                              max =          9
Estimated covariances      =      2485          R-squared         =     0.9116
Estimated autocorrelations =         0          Wald chi2(17)     =    5572.94
Estimated coefficients     =        36          Prob > chi2       =     0.0000

-----------------------------------------------------------------------------------
                  |           Panel-corrected
         totalpct |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
       1.timezone |   3.786182   1.191224     3.18   0.001     1.451426    6.120938
                  |
         agegroup |
               2  |   6.937542   .5926924    11.71   0.000     5.775887    8.099198
               3  |   19.89415    1.23791    16.07   0.000     17.46789    22.32041
               4  |   31.63006   1.487938    21.26   0.000     28.71376    34.54637
               5  |   34.57573   1.386212    24.94   0.000      31.8588    37.29265
                  |
timezone#agegroup |
             1 2  |  -.8786776   .1953032    -4.50   0.000    -1.261465   -.4958905
             1 3  |  -2.161084   .5014238    -4.31   0.000    -3.143857   -1.178312
             1 4  |  -4.501023   .3767534   -11.95   0.000    -5.239446     -3.7626
             1 5  |  -6.054107   .3970033   -15.25   0.000    -6.832219   -5.275995
                  |
         congdist |
               2  |  -8.802208   2.109907    -4.17   0.000    -12.93755   -4.666866
               5  |  -6.432167    1.80449    -3.56   0.000    -9.968901   -2.895432
                  |
             year |
            1999  |   -44.9022   2.226831   -20.16   0.000     -49.2667   -40.53769
            2003  |   -22.1796   1.667502   -13.30   0.000    -25.44784   -18.91135
            2006  |  -2.087483   1.667502    -1.25   0.211    -5.355728    1.180761
            2007  |  -25.07796   1.667502   -15.04   0.000    -28.34621   -21.80972
            2010  |  -6.551336   1.667502    -3.93   0.000     -9.81958   -3.283091
            2011  |  -30.69374   1.667502   -18.41   0.000    -33.96199    -27.4255
            2014  |   -7.62819   1.855729    -4.11   0.000    -11.26535   -3.991027
            2015  |  -28.15959   1.855729   -15.17   0.000    -31.79675   -24.52243
                  |
    congdist#year |
          2 1999  |   19.56359   2.969108     6.59   0.000     13.74424    25.38293
          2 2003  |   9.692131    2.25855     4.29   0.000     5.265453    14.11881
          2 2006  |   3.807045    2.25855     1.69   0.092    -.6196321    8.233723
          2 2007  |   10.46433    2.25855     4.63   0.000      6.03765      14.891
          2 2010  |    5.86945    2.25855     2.60   0.009     1.442772    10.29613
          2 2011  |   8.194351    2.25855     3.63   0.000     3.767673    12.62103
          2 2014  |   1.783264   2.686638     0.66   0.507    -3.482449    7.048978
          2 2015  |   7.414765   2.686638     2.76   0.006     2.149051    12.68048
          5 1999  |   13.23129   2.226831     5.94   0.000     8.866786     17.5958
          5 2003  |    6.76366   1.667502     4.06   0.000     3.495415     10.0319
          5 2006  |  -.2969156   1.667502    -0.18   0.859     -3.56516    2.971329
          5 2007  |   7.487404   1.667502     4.49   0.000      4.21916    10.75565
          5 2010  |   4.675524   1.667502     2.80   0.005      1.40728    7.943769
          5 2011  |    4.42388   1.667502     2.65   0.008     1.155635    7.692124
          5 2014  |   1.454671   1.855729     0.78   0.433    -2.182491    5.091834
          5 2015  |   3.921341   1.855729     2.11   0.035     .2841782    7.558504
                  |
            _cons |   38.05292   2.208308    17.23   0.000     33.72472    42.38113
-----------------------------------------------------------------------------------

. 
. 
. 
. glm totalprop timezone agegroup interact congdist##year if mod(year, 4) == 0 & expansion == 1, cl(countyyear)
>  family(binomial) 
note: totalprop has noninteger values

Iteration 0:   log pseudolikelihood = -154.50074  
Iteration 1:   log pseudolikelihood =  -154.4974  
Iteration 2:   log pseudolikelihood =  -154.4974  

Generalized linear models                         No. of obs      =        350
Optimization     : ML                             Residual df     =        332
                                                  Scale parameter =          1
Deviance         =  3.346351419                   (1/df) Deviance =   .0100794
Pearson          =  3.349496129                   (1/df) Pearson  =   .0100888

Variance function: V(u) = u*(1-u/1)               [Binomial]
Link function    : g(u) = ln(u/(1-u))             [Logit]

                                                  AIC             =   .9856994
Log pseudolikelihood = -154.4973978               BIC             =  -1941.487

                             (Std. Err. adjusted for 70 clusters in countyyear)
-------------------------------------------------------------------------------
              |               Robust
    totalprop |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |   .0796858   .0508996     1.57   0.117    -.0200756    .1794472
     agegroup |   .2598688   .0121861    21.32   0.000     .2359844    .2837533
     interact |  -.0526125    .015378    -3.42   0.001    -.0827528   -.0224722
              |
     congdist |
           2  |  -.0956418   .0599458    -1.60   0.111    -.2131335    .0218498
           5  |  -.3060513   .1505626    -2.03   0.042    -.6011486    -.010954
              |
         year |
        2004  |   .1107924    .041901     2.64   0.008     .0286679    .1929168
        2008  |    .072692   .0537234     1.35   0.176    -.0326039    .1779879
        2012  |   .0419248   .0534744     0.78   0.433    -.0628831    .1467326
        2016  |   .1495558   .0466164     3.21   0.001     .0581893    .2409223
              |
congdist#year |
      2 2004  |   .1774733   .0711139     2.50   0.013     .0380927     .316854
      2 2008  |   .1756671   .0835221     2.10   0.035     .0119669    .3393674
      2 2012  |  -.0738292   .0740361    -1.00   0.319    -.2189373    .0712789
      2 2016  |  -.0133797   .0698082    -0.19   0.848    -.1502013    .1234419
      5 2004  |   .2098304   .1762667     1.19   0.234     -.135646    .5553068
      5 2008  |   .1733212   .1990745     0.87   0.384    -.2168577    .5635001
      5 2012  |   .0875848   .1859378     0.47   0.638    -.2768465    .4520161
      5 2016  |   .0840435   .1710866     0.49   0.623      -.25128     .419367
              |
        _cons |  -.5015154   .0561546    -8.93   0.000    -.6115764   -.3914543
-------------------------------------------------------------------------------

. 
. glm totalprop timezone agegroup interact congdist##year if mod(year, 4) != 0 & expansion == 1, cl(countyyear)
>  family(binomial) 
note: totalprop has noninteger values

Iteration 0:   log pseudolikelihood = -253.75308  
Iteration 1:   log pseudolikelihood = -253.31009  
Iteration 2:   log pseudolikelihood = -253.30848  
Iteration 3:   log pseudolikelihood = -253.30848  

Generalized linear models                         No. of obs      =        630
Optimization     : ML                             Residual df     =        600
                                                  Scale parameter =          1
Deviance         =  9.346827557                   (1/df) Deviance =    .015578
Pearson          =  9.346266114                   (1/df) Pearson  =   .0155771

Variance function: V(u) = u*(1-u/1)               [Binomial]
Link function    : g(u) = ln(u/(1-u))             [Logit]

                                                  AIC             =    .899392
Log pseudolikelihood = -253.3084814               BIC             =  -3858.085

                            (Std. Err. adjusted for 126 clusters in countyyear)
-------------------------------------------------------------------------------
              |               Robust
    totalprop |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |   .3079966   .0694051     4.44   0.000     .1719652     .444028
     agegroup |   .4470953   .0124889    35.80   0.000     .4226176    .4715731
     interact |  -.0809412   .0154092    -5.25   0.000    -.1111426   -.0507398
              |
     congdist |
           2  |  -.3687395   .1565731    -2.36   0.019    -.6756172   -.0618619
           5  |  -.2569217   .1654955    -1.55   0.121    -.5812869    .0674435
              |
         year |
        1999  |  -2.346357   .2639819    -8.89   0.000    -2.863752   -1.828962
        2003  |  -.9665743   .1515516    -6.38   0.000     -1.26361   -.6695385
        2006  |  -.0844167    .207637    -0.41   0.684    -.4913777    .3225443
        2007  |  -1.104654   .1585597    -6.97   0.000    -1.415425   -.7938826
        2010  |  -.2771015   .1649893    -1.68   0.093    -.6004746    .0462715
        2011  |  -1.391383   .1370321   -10.15   0.000    -1.659961   -1.122806
        2014  |  -.3206563   .1492439    -2.15   0.032    -.6131689   -.0281437
        2015  |  -1.260774   .1452804    -8.68   0.000    -1.545518   -.9760293
              |
congdist#year |
      2 1999  |   1.119897   .2988482     3.75   0.000     .5341656    1.705629
      2 2003  |   .4094392     .17923     2.28   0.022     .0581549    .7607235
      2 2006  |   .1579563   .2353447     0.67   0.502    -.3033108    .6192234
      2 2007  |   .4468967   .1849188     2.42   0.016     .0844626    .8093308
      2 2010  |   .2465984   .2065583     1.19   0.233    -.1582484    .6514452
      2 2011  |   .3296402   .1703144     1.94   0.053    -.0041698    .6634502
      2 2014  |   .0638594   .1787731     0.36   0.721    -.2865294    .4142482
      2 2015  |   .2956691   .1770322     1.67   0.095    -.0513075    .6426458
      5 1999  |   .7346327    .330564     2.22   0.026     .0867391    1.382526
      5 2003  |   .2695046   .2163033     1.25   0.213    -.1544421    .6934514
      5 2006  |  -.0204149   .2509951    -0.08   0.935    -.5123562    .4715264
      5 2007  |   .3017647   .1861027     1.62   0.105    -.0629898    .6665192
      5 2010  |   .1946483   .2584294     0.75   0.451     -.311864    .7011607
      5 2011  |   .1244555   .1703229     0.73   0.465    -.2093712    .4582822
      5 2014  |   .0482498   .1817993     0.27   0.791    -.3080702    .4045698
      5 2015  |   .1101677   .2074621     0.53   0.595    -.2964504    .5167859
              |
        _cons |  -1.074334   .1391766    -7.72   0.000    -1.347115    -.801553
-------------------------------------------------------------------------------

. 
. 
. 
. glm totalprop timezone##agegroup congdist##year if mod(year, 4) == 0 & expansion == 1, cl(countyyear) family(
> binomial) 
note: totalprop has noninteger values

Iteration 0:   log pseudolikelihood =  -153.8328  
Iteration 1:   log pseudolikelihood = -153.82979  
Iteration 2:   log pseudolikelihood = -153.82979  

Generalized linear models                         No. of obs      =        350
Optimization     : ML                             Residual df     =        326
                                                  Scale parameter =          1
Deviance         =  2.011138936                   (1/df) Deviance =   .0061691
Pearson          =  2.010047977                   (1/df) Pearson  =   .0061658

Variance function: V(u) = u*(1-u/1)               [Binomial]
Link function    : g(u) = ln(u/(1-u))             [Logit]

                                                  AIC             =    1.01617
Log pseudolikelihood = -153.8297915               BIC             =  -1907.675

                                 (Std. Err. adjusted for 70 clusters in countyyear)
-----------------------------------------------------------------------------------
                  |               Robust
        totalprop |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
       1.timezone |   .0185809   .0408954     0.45   0.650    -.0615726    .0987343
                  |
         agegroup |
               2  |   .0719641   .0226029     3.18   0.001     .0276633    .1162649
               3  |   .5331785   .0282746    18.86   0.000     .4777613    .5885958
               4  |   .9181805   .0364413    25.20   0.000     .8467568    .9896042
               5  |   .8684627   .0488532    17.78   0.000     .7727121    .9642132
                  |
timezone#agegroup |
             1 2  |  -.0479716    .033134    -1.45   0.148     -.112913    .0169699
             1 3  |  -.0846107   .0387887    -2.18   0.029    -.1606351   -.0085863
             1 4  |  -.1328175   .0493895    -2.69   0.007    -.2296192   -.0360159
             1 5  |  -.2176713   .0628618    -3.46   0.001    -.3408781   -.0944645
                  |
         congdist |
               2  |  -.0960095   .0601914    -1.60   0.111    -.2139825    .0219636
               5  |  -.3071536   .1511124    -2.03   0.042    -.6033285   -.0109787
                  |
             year |
            2004  |   .1112556   .0420798     2.64   0.008     .0287806    .1937305
            2008  |   .0729981   .0539533     1.35   0.176    -.0327485    .1787446
            2012  |   .0421033   .0536969     0.78   0.433    -.0631406    .1473473
            2016  |   .1501934   .0468076     3.21   0.001     .0584521    .2419346
                  |
    congdist#year |
          2 2004  |   .1781561   .0714013     2.50   0.013     .0382121    .3181001
          2 2008  |    .176349   .0838621     2.10   0.035     .0119823    .3407156
          2 2012  |  -.0741459    .074343    -1.00   0.319    -.2198555    .0715637
          2 2016  |  -.0134768   .0700938    -0.19   0.848     -.150858    .1239045
          5 2004  |   .2105452   .1769136     1.19   0.234    -.1361991    .5572894
          5 2008  |   .1739157     .19982     0.87   0.384    -.2177243    .5655556
          5 2012  |   .0878762   .1866209     0.47   0.638     -.277894    .4536464
          5 2016  |   .0842604   .1717087     0.49   0.624    -.2522824    .4208032
                  |
            _cons |  -.1997989   .0511857    -3.90   0.000    -.3001211   -.0994767
-----------------------------------------------------------------------------------

. 
. glm totalprop timezone##agegroup congdist##year if mod(year, 4) != 0 & expansion == 1, cl(countyyear) family(
> binomial) 
note: totalprop has noninteger values

Iteration 0:   log pseudolikelihood = -252.82551  
Iteration 1:   log pseudolikelihood = -252.31286  
Iteration 2:   log pseudolikelihood = -252.31084  
Iteration 3:   log pseudolikelihood = -252.31084  

Generalized linear models                         No. of obs      =        630
Optimization     : ML                             Residual df     =        594
                                                  Scale parameter =          1
Deviance         =  7.351537751                   (1/df) Deviance =   .0123763
Pearson          =  7.367827681                   (1/df) Pearson  =   .0124038

Variance function: V(u) = u*(1-u/1)               [Binomial]
Link function    : g(u) = ln(u/(1-u))             [Logit]

                                                  AIC             =   .9152725
Log pseudolikelihood = -252.3108365               BIC             =  -3821.406

                                (Std. Err. adjusted for 126 clusters in countyyear)
-----------------------------------------------------------------------------------
                  |               Robust
        totalprop |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
       1.timezone |   .2367677   .0590561     4.01   0.000     .1210199    .3525155
                  |
         agegroup |
               2  |    .426209   .0192679    22.12   0.000     .3884445    .4639734
               3  |   1.062212   .0274974    38.63   0.000     1.008318    1.116106
               4  |   1.570202   .0379977    41.32   0.000     1.495728    1.644676
               5  |   1.695677   .0516205    32.85   0.000     1.594503    1.796851
                  |
timezone#agegroup |
             1 2  |  -.0878438   .0262913    -3.34   0.001    -.1393738   -.0363137
             1 3  |  -.1662279   .0348797    -4.77   0.000    -.2345909   -.0978649
             1 4  |   -.262946    .048177    -5.46   0.000    -.3573711   -.1685208
             1 5  |  -.3281276   .0648707    -5.06   0.000    -.4552718   -.2009834
                  |
         congdist |
               2  |  -.3709794    .157553    -2.35   0.019    -.6797777   -.0621811
               5  |   -.258416   .1665765    -1.55   0.121    -.5848999    .0680679
                  |
             year |
            1999  |  -2.349725   .2640231    -8.90   0.000    -2.867201    -1.83225
            2003  |  -.9709214   .1523165    -6.37   0.000    -1.269456   -.6723866
            2006  |  -.0848664   .2090235    -0.41   0.685    -.4945451    .3248122
            2007  |  -1.109171   .1592303    -6.97   0.000    -1.421257   -.7970856
            2010  |  -.2788014   .1660345    -1.68   0.093    -.6042231    .0466203
            2011  |  -1.395933   .1377547   -10.13   0.000    -1.665927   -1.125938
            2014  |  -.3226094   .1502115    -2.15   0.032    -.6170186   -.0282003
            2015  |  -1.265356   .1459803    -8.67   0.000    -1.551472   -.9792399
                  |
    congdist#year |
          2 1999  |   1.121294     .29896     3.75   0.000     .5353436    1.707245
          2 2003  |   .4116831   .1801373     2.29   0.022     .0586205    .7647457
          2 2006  |   .1588513   .2368776     0.67   0.502    -.3054203    .6231228
          2 2007  |   .4491657   .1857543     2.42   0.016      .085094    .8132374
          2 2010  |   .2481283    .207819     1.19   0.232    -.1591894    .6554459
          2 2011  |    .331963   .1711619     1.94   0.052    -.0035082    .6674342
          2 2014  |   .0645847   .1798804     0.36   0.720    -.2879744    .4171439
          2 2015  |   .2979357   .1778827     1.67   0.094    -.0507081    .6465794
          5 1999  |   .7363041   .3305026     2.23   0.026      .088531    1.384077
          5 2003  |   .2708229   .2172303     1.25   0.213    -.1549407    .6965864
          5 2006  |  -.0206889   .2526809    -0.08   0.935    -.5159344    .4745566
          5 2007  |   .3031568   .1870273     1.62   0.105      -.06341    .6697237
          5 2010  |   .1957715   .2601432     0.75   0.452    -.3140997    .7056428
          5 2011  |   .1263425   .1712899     0.74   0.461    -.2093795    .4620645
          5 2014  |   .0485136   .1829976     0.27   0.791    -.3101551    .4071823
          5 2015  |   .1118583   .2081868     0.54   0.591    -.2961804     .519897
                  |
            _cons |  -.6880097   .1372099    -5.01   0.000    -.9569361   -.4190832
-----------------------------------------------------------------------------------

. 
. 
. 
. 
. 
. * Appendix: Table 1 replicated including counties near but not touching time-zone border

. 
. 
. 
. xtpcse totalpct timezone agegroup interact congdist##year

Number of gaps in sample:  460
note: 6.congdist#2012.year identifies no observations in the sample
note: 6.congdist#2014.year identifies no observations in the sample
note: 6.congdist#2015.year identifies no observations in the sample
note: 6.congdist#2016.year identifies no observations in the sample

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =      1,610
Time variable:    year                          Number of groups  =        115
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =         14
                                                              avg =         14
                                                              max =         14
Estimated covariances      =      6670          R-squared         =     0.8903
Estimated autocorrelations =         0          Wald chi2(27)     =    3360.75
Estimated coefficients     =        55          Prob > chi2       =     0.0000

-------------------------------------------------------------------------------
              |           Panel-corrected
     totalpct |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |   3.877469   .9607051     4.04   0.000     1.994522    5.760416
     agegroup |   8.184616   .4834384    16.93   0.000     7.237094    9.132138
     interact |  -1.241709     .08643   -14.37   0.000    -1.411109    -1.07231
              |
     congdist |
           2  |  -8.163217   1.437433    -5.68   0.000    -10.98053     -5.3459
           5  |  -9.235669   1.225351    -7.54   0.000    -11.63731   -6.834024
           6  |  -17.22924   2.181347    -7.90   0.000     -21.5046   -12.95387
              |
         year |
        1999  |  -42.83647   1.580267   -27.11   0.000    -45.93374   -39.73921
        2000  |  -2.807777   1.580267    -1.78   0.076    -5.905043    .2894884
        2003  |   -23.2911   1.217594   -19.13   0.000    -25.67754   -20.90466
        2004  |   .3863403   1.217594     0.32   0.751    -2.000099     2.77278
        2006  |  -1.732553   1.217594    -1.42   0.155    -4.118993     .653886
        2007  |  -25.89922   1.217594   -21.27   0.000    -28.28566   -23.51278
        2008  |  -1.119894   1.217594    -0.92   0.358    -3.506333    1.266546
        2010  |  -4.708239   1.217594    -3.87   0.000    -7.094678     -2.3218
        2011  |  -29.07987   1.217594   -23.88   0.000    -31.46631   -26.69343
        2012  |  -1.847036   1.326765    -1.39   0.164    -4.447448    .7533763
        2014  |  -7.576564   1.326765    -5.71   0.000    -10.17698   -4.976151
        2015  |  -29.54719   1.326765   -22.27   0.000     -32.1476   -26.94678
        2016  |   .8881751   1.449855     0.61   0.540    -1.953489    3.729839
              |
congdist#year |
      2 1999  |   15.06924   2.001671     7.53   0.000     11.14604    18.99245
      2 2000  |   7.977132   2.001671     3.99   0.000     4.053929    11.90034
      2 2003  |   10.35991   1.579814     6.56   0.000     7.263528    13.45628
      2 2004  |   11.44413   1.579814     7.24   0.000     8.347756    14.54051
      2 2006  |   2.122027   1.579814     1.34   0.179    -.9743521    5.218405
      2 2007  |   12.17077   1.579814     7.70   0.000     9.074387    15.26714
      2 2008  |   12.59987   1.579814     7.98   0.000     9.503494    15.69625
      2 2010  |   2.918498   1.579814     1.85   0.065    -.1778806    6.014876
      2 2011  |   5.865258   1.579814     3.71   0.000     2.768879    8.961637
      2 2012  |   6.736307   1.870172     3.60   0.000     3.070837    10.40178
      2 2014  |    2.35609   1.870172     1.26   0.208     -1.30938    6.021561
      2 2015  |   7.912603   1.870172     4.23   0.000     4.247133    11.57807
      2 2016  |   7.551775   1.953566     3.87   0.000     3.722856    11.38069
      5 1999  |   10.32464   1.288482     8.01   0.000     7.799263    12.85002
      5 2000  |   .7632734   1.288482     0.59   0.554    -1.762105    3.288651
      5 2003  |    6.33206    1.01351     6.25   0.000     4.345617    8.318503
      5 2004  |   5.143142    1.01351     5.07   0.000     3.156699    7.129585
      5 2006  |   1.654882    1.01351     1.63   0.103    -.3315614    3.641325
      5 2007  |   7.148382    1.01351     7.05   0.000     5.161939    9.134825
      5 2008  |   5.884215    1.01351     5.81   0.000     3.897772    7.870658
      5 2010  |   4.658978    1.01351     4.60   0.000     2.672535    6.645421
      5 2011  |    2.49345    1.01351     2.46   0.014     .5070069    4.479893
      5 2012  |    5.25185   1.164923     4.51   0.000     2.968643    7.535056
      5 2014  |   2.273176   1.164923     1.95   0.051    -.0100308    4.556383
      5 2015  |   5.384429   1.164923     4.62   0.000     3.101222    7.667636
      5 2016  |   4.129734    1.30013     3.18   0.001     1.581526    6.677941
      6 1999  |     16.401   2.915386     5.63   0.000     10.68694    22.11505
      6 2000  |   11.71099   2.915386     4.02   0.000     5.996941    17.42505
      6 2003  |   15.87495   2.517855     6.30   0.000     10.94005    20.80985
      6 2004  |   14.38263   2.517855     5.71   0.000     9.447727    19.31754
      6 2006  |   2.660218   2.517855     1.06   0.291    -2.274687    7.595122
      6 2007  |   15.74604   2.517855     6.25   0.000     10.81113    20.68094
      6 2008  |   14.48284   2.517855     5.75   0.000     9.547935    19.41774
      6 2010  |   9.080507   2.517855     3.61   0.000     4.145602    14.01541
      6 2011  |   11.61179   2.517855     4.61   0.000     6.676884    16.54669
      6 2012  |          0  (empty)
      6 2014  |          0  (empty)
      6 2015  |          0  (empty)
      6 2016  |          0  (empty)
              |
        _cons |   32.76653   2.253213    14.54   0.000     28.35032    37.18275
-------------------------------------------------------------------------------

. 
. xtpcse totalpct timezone##agegroup congdist##year

Number of gaps in sample:  460
note: 6.congdist#2012.year identifies no observations in the sample
note: 6.congdist#2014.year identifies no observations in the sample
note: 6.congdist#2015.year identifies no observations in the sample
note: 6.congdist#2016.year identifies no observations in the sample

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =      1,610
Time variable:    year                          Number of groups  =        115
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =         14
                                                              avg =         14
                                                              max =         14
Estimated covariances      =      6670          R-squared         =     0.9098
Estimated autocorrelations =         0          Wald chi2(33)     =   12221.09
Estimated coefficients     =        61          Prob > chi2       =     0.0000

-----------------------------------------------------------------------------------
                  |           Panel-corrected
         totalpct |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
       1.timezone |   2.272508   .9005843     2.52   0.012     .5073952    4.037621
                  |
         agegroup |
               2  |   5.399665    .819944     6.59   0.000     3.792604    7.006725
               3  |   17.32133   1.244937    13.91   0.000      14.8813    19.76136
               4  |   27.86062   1.636944    17.02   0.000     24.65227    31.06897
               5  |    29.6926   2.029962    14.63   0.000     25.71395    33.67125
                  |
timezone#agegroup |
             1 2  |  -.8882073   .2236848    -3.97   0.000    -1.326621   -.4497931
             1 3  |  -1.456701     .33733    -4.32   0.000    -2.117856   -.7955467
             1 4  |   -3.20655   .3703414    -8.66   0.000    -3.932405   -2.480694
             1 5  |  -5.049375    .287151   -17.58   0.000     -5.61218   -4.486569
                  |
         congdist |
               2  |  -8.163217   1.437433    -5.68   0.000    -10.98053     -5.3459
               5  |  -9.235669   1.225351    -7.54   0.000    -11.63731   -6.834024
               6  |  -17.22924   2.181347    -7.90   0.000     -21.5046   -12.95387
                  |
             year |
            1999  |  -42.83647   1.580267   -27.11   0.000    -45.93374   -39.73921
            2000  |  -2.807777   1.580267    -1.78   0.076    -5.905043    .2894884
            2003  |   -23.2911   1.217594   -19.13   0.000    -25.67754   -20.90466
            2004  |   .3863403   1.217594     0.32   0.751    -2.000099     2.77278
            2006  |  -1.732553   1.217594    -1.42   0.155    -4.118993     .653886
            2007  |  -25.89922   1.217594   -21.27   0.000    -28.28566   -23.51278
            2008  |  -1.119894   1.217594    -0.92   0.358    -3.506333    1.266546
            2010  |  -4.708239   1.217594    -3.87   0.000    -7.094678     -2.3218
            2011  |  -29.07987   1.217594   -23.88   0.000    -31.46631   -26.69343
            2012  |  -1.847036   1.326765    -1.39   0.164    -4.447448    .7533763
            2014  |  -7.576564   1.326765    -5.71   0.000    -10.17698   -4.976151
            2015  |  -29.54719   1.326765   -22.27   0.000     -32.1476   -26.94678
            2016  |   .8881751   1.449855     0.61   0.540    -1.953489    3.729839
                  |
    congdist#year |
          2 1999  |   15.06924   2.001671     7.53   0.000     11.14604    18.99245
          2 2000  |   7.977132   2.001671     3.99   0.000     4.053929    11.90034
          2 2003  |   10.35991   1.579814     6.56   0.000     7.263528    13.45628
          2 2004  |   11.44413   1.579814     7.24   0.000     8.347756    14.54051
          2 2006  |   2.122027   1.579814     1.34   0.179    -.9743521    5.218405
          2 2007  |   12.17077   1.579814     7.70   0.000     9.074387    15.26714
          2 2008  |   12.59987   1.579814     7.98   0.000     9.503494    15.69625
          2 2010  |   2.918498   1.579814     1.85   0.065    -.1778806    6.014876
          2 2011  |   5.865258   1.579814     3.71   0.000     2.768879    8.961637
          2 2012  |   6.736307   1.870172     3.60   0.000     3.070837    10.40178
          2 2014  |    2.35609   1.870172     1.26   0.208     -1.30938    6.021561
          2 2015  |   7.912603   1.870172     4.23   0.000     4.247133    11.57807
          2 2016  |   7.551775   1.953566     3.87   0.000     3.722856    11.38069
          5 1999  |   10.32464   1.288482     8.01   0.000     7.799263    12.85002
          5 2000  |   .7632734   1.288482     0.59   0.554    -1.762105    3.288651
          5 2003  |    6.33206    1.01351     6.25   0.000     4.345617    8.318503
          5 2004  |   5.143142    1.01351     5.07   0.000     3.156699    7.129585
          5 2006  |   1.654882    1.01351     1.63   0.103    -.3315614    3.641325
          5 2007  |   7.148382    1.01351     7.05   0.000     5.161939    9.134825
          5 2008  |   5.884215    1.01351     5.81   0.000     3.897772    7.870658
          5 2010  |   4.658978    1.01351     4.60   0.000     2.672535    6.645421
          5 2011  |    2.49345    1.01351     2.46   0.014     .5070069    4.479893
          5 2012  |    5.25185   1.164923     4.51   0.000     2.968643    7.535056
          5 2014  |   2.273176   1.164923     1.95   0.051    -.0100308    4.556383
          5 2015  |   5.384429   1.164923     4.62   0.000     3.101222    7.667636
          5 2016  |   4.129734    1.30013     3.18   0.001     1.581526    6.677941
          6 1999  |     16.401   2.915386     5.63   0.000     10.68694    22.11505
          6 2000  |   11.71099   2.915386     4.02   0.000     5.996941    17.42505
          6 2003  |   15.87495   2.517855     6.30   0.000     10.94005    20.80985
          6 2004  |   14.38263   2.517855     5.71   0.000     9.447727    19.31754
          6 2006  |   2.660218   2.517855     1.06   0.291    -2.274687    7.595122
          6 2007  |   15.74604   2.517855     6.25   0.000     10.81113    20.68094
          6 2008  |   14.48284   2.517855     5.75   0.000     9.547935    19.41774
          6 2010  |   9.080507   2.517855     3.61   0.000     4.145602    14.01541
          6 2011  |   11.61179   2.517855     4.61   0.000     6.676884    16.54669
          6 2012  |          0  (empty)
          6 2014  |          0  (empty)
          6 2015  |          0  (empty)
          6 2016  |          0  (empty)
                  |
            _cons |   41.26554   2.019524    20.43   0.000     37.30734    45.22373
-----------------------------------------------------------------------------------

. 
. glm totalprop timezone agegroup interact congdist##year, cl(countyyear) family(binomial)
note: 6.congdist#2012.year identifies no observations in the sample
note: 6.congdist#2014.year identifies no observations in the sample
note: 6.congdist#2015.year identifies no observations in the sample
note: 6.congdist#2016.year identifies no observations in the sample
note: totalprop has noninteger values

Iteration 0:   log pseudolikelihood = -671.21172  
Iteration 1:   log pseudolikelihood = -670.65528  
Iteration 2:   log pseudolikelihood = -670.65388  
Iteration 3:   log pseudolikelihood = -670.65388  

Generalized linear models                         No. of obs      =      1,610
Optimization     : ML                             Residual df     =      1,565
                                                  Scale parameter =          1
Deviance         =  27.28194211                   (1/df) Deviance =   .0174326
Pearson          =  27.13240103                   (1/df) Pearson  =    .017337

Variance function: V(u) = u*(1-u/1)               [Binomial]
Link function    : g(u) = ln(u/(1-u))             [Logit]

                                                  AIC             =    .889011
Log pseudolikelihood = -670.6538755               BIC             =  -11528.66

                            (Std. Err. adjusted for 322 clusters in countyyear)
-------------------------------------------------------------------------------
              |               Robust
    totalprop |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |   .1949852   .0465864     4.19   0.000     .1036776    .2862928
     agegroup |   .3722131   .0100495    37.04   0.000     .3525164    .3919098
     interact |  -.0615838   .0127596    -4.83   0.000    -.0865921   -.0365754
              |
     congdist |
           2  |  -.3429784   .1152441    -2.98   0.003    -.5688526   -.1171042
           5  |  -.3871593   .1379366    -2.81   0.005    -.6575101   -.1168086
           6  |  -.7340994   .0967319    -7.59   0.000    -.9236905   -.5445082
              |
         year |
        1999  |  -2.146242   .2215859    -9.69   0.000    -2.580542   -1.711941
        2000  |  -.1194025   .0981336    -1.22   0.224    -.3117408    .0729357
        2003  |  -1.002627   .1143535    -8.77   0.000    -1.226756   -.7784984
        2004  |   .0182198   .0950273     0.19   0.848    -.1680304    .2044699
        2006  |   -.072483   .1547817    -0.47   0.640    -.3758496    .2308836
        2007  |  -1.126232   .1183588    -9.52   0.000    -1.358211    -.894253
        2008  |  -.0463583   .0998798    -0.46   0.643    -.2421192    .1494026
        2010  |  -.1985122   .1279575    -1.55   0.121    -.4493044      .05228
        2011  |  -1.284407   .1259656   -10.20   0.000    -1.531295   -1.037519
        2012  |   -.075642   .0994013    -0.76   0.447    -.2704649     .119181
        2014  |  -.3184932   .1177821    -2.70   0.007    -.5493418   -.0876445
        2015  |  -1.311764   .1101372   -11.91   0.000    -1.527629   -1.095899
        2016  |   .0398618   .0982831     0.41   0.685    -.1527695     .232493
              |
congdist#year |
      2 1999  |   .8164326   .2507442     3.26   0.001      .324983    1.307882
      2 2000  |   .3385257   .1230323     2.75   0.006     .0973869    .5796645
      2 2003  |   .4413847   .1371073     3.22   0.001     .1726595      .71011
      2 2004  |   .4895865   .1208609     4.05   0.000     .2527035    .7264694
      2 2006  |   .0889418   .1789447     0.50   0.619    -.2617832    .4396669
      2 2007  |   .5285242   .1397838     3.78   0.000      .254553    .8024954
      2 2008  |   .5386425   .1281652     4.20   0.000     .2874432    .7898417
      2 2010  |   .1227125   .1550792     0.79   0.429    -.1812372    .4266623
      2 2011  |   .2170497   .1455731     1.49   0.136    -.0682684    .5023677
      2 2012  |   .2821007   .1289326     2.19   0.029     .0293975    .5348039
      2 2014  |   .0966533   .1430366     0.68   0.499    -.1836932    .3769998
      2 2015  |   .3301071   .1349203     2.45   0.014     .0656681     .594546
      2 2016  |   .3195433   .1222701     2.61   0.009     .0798983    .5591882
      5 1999  |   .4441604   .3353358     1.32   0.185    -.2130857    1.101406
      5 2000  |   .0317169   .2057191     0.15   0.877    -.3714851    .4349189
      5 2003  |   .2386192   .2271279     1.05   0.293    -.2065432    .6837816
      5 2004  |   .2184257   .1928967     1.13   0.257    -.1596448    .5964963
      5 2006  |   .0691582   .1948265     0.35   0.723    -.3126948    .4510111
      5 2007  |    .272378   .2063812     1.32   0.187    -.1321217    .6768777
      5 2008  |   .2501688   .1853192     1.35   0.177    -.1130503    .6133878
      5 2010  |   .1964036   .1967343     1.00   0.318    -.1891887    .5819958
      5 2011  |  -.0065738   .2034647    -0.03   0.974    -.4053573    .3922097
      5 2012  |    .221235   .1515825     1.46   0.144    -.0758613    .5183313
      5 2014  |   .0898146   .1593653     0.56   0.573    -.2225357    .4021648
      5 2015  |   .1655044   .1933413     0.86   0.392    -.2134377    .5444465
      5 2016  |   .1748241   .1508081     1.16   0.246    -.1207543    .4704026
      6 1999  |   .6321599   .2216184     2.85   0.004     .1977959    1.066524
      6 2000  |   .5052573    .098142     5.15   0.000     .3129025    .6976122
      6 2003  |   .6611084   .1143563     5.78   0.000     .4369743    .8852426
      6 2004  |   .6189843   .0950729     6.51   0.000     .4326447    .8053238
      6 2006  |   .1135049   .1547798     0.73   0.463    -.1898579    .4168676
      6 2007  |   .6499298   .1183623     5.49   0.000     .4179441    .8819156
      6 2008  |   .6230683   .0998733     6.24   0.000     .4273203    .8188164
      6 2010  |   .3897896   .1279631     3.05   0.002     .1389865    .6405926
      6 2011  |   .4088257   .1259688     3.25   0.001     .1619315      .65572
      6 2012  |          0  (empty)
      6 2014  |          0  (empty)
      6 2015  |          0  (empty)
      6 2016  |          0  (empty)
              |
        _cons |   -.811386   .1019826    -7.96   0.000    -1.011268   -.6115038
-------------------------------------------------------------------------------

. 
. glm totalprop timezone##agegroup congdist##year, cl(countyyear) family(binomial)
note: 6.congdist#2012.year identifies no observations in the sample
note: 6.congdist#2014.year identifies no observations in the sample
note: 6.congdist#2015.year identifies no observations in the sample
note: 6.congdist#2016.year identifies no observations in the sample
note: totalprop has noninteger values

Iteration 0:   log pseudolikelihood = -668.94958  
Iteration 1:   log pseudolikelihood = -668.36681  
Iteration 2:   log pseudolikelihood = -668.36535  
Iteration 3:   log pseudolikelihood = -668.36535  

Generalized linear models                         No. of obs      =      1,610
Optimization     : ML                             Residual df     =      1,559
                                                  Scale parameter =          1
Deviance         =  22.70488213                   (1/df) Deviance =   .0145637
Pearson          =  22.45378361                   (1/df) Pearson  =   .0144027

Variance function: V(u) = u*(1-u/1)               [Binomial]
Link function    : g(u) = ln(u/(1-u))             [Logit]

                                                  AIC             =   .8936215
Log pseudolikelihood = -668.3653455               BIC             =  -11488.93

                                (Std. Err. adjusted for 322 clusters in countyyear)
-----------------------------------------------------------------------------------
                  |               Robust
        totalprop |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
       1.timezone |   .1257095   .0403099     3.12   0.002     .0467034    .2047155
                  |
         agegroup |
               2  |   .2749574   .0185585    14.82   0.000     .2385835    .3113314
               3  |   .8200981   .0265203    30.92   0.000     .7681193    .8720769
               4  |   1.280132   .0335248    38.18   0.000     1.214424    1.345839
               5  |   1.361165   .0428462    31.77   0.000     1.277188    1.445142
                  |
timezone#agegroup |
             1 2  |  -.0578301    .025383    -2.28   0.023    -.1075798   -.0080804
             1 3  |  -.0948075   .0346251    -2.74   0.006    -.1626714   -.0269436
             1 4  |   -.170564   .0427386    -3.99   0.000      -.25433   -.0867979
             1 5  |  -.2520774   .0552715    -4.56   0.000    -.3604076   -.1437472
                  |
         congdist |
               2  |  -.3444946     .11574    -2.98   0.003    -.5713409   -.1176483
               5  |  -.3889152   .1384894    -2.81   0.005    -.6603494    -.117481
               6  |  -.7368596   .0971601    -7.58   0.000    -.9272899   -.5464292
                  |
             year |
            1999  |      -2.15   .2216635    -9.70   0.000    -2.584453   -1.715548
            2000  |  -.1199334   .0985754    -1.22   0.224    -.3131376    .0732709
            2003  |  -1.005922    .114757    -8.77   0.000    -1.230842   -.7810025
            2004  |   .0182784   .0954562     0.19   0.848    -.1688124    .2053692
            2006  |  -.0728334    .155487    -0.47   0.639    -.3775823    .2319155
            2007  |  -1.129706    .118734    -9.51   0.000     -1.36242   -.8969911
            2008  |  -.0465922   .1003319    -0.46   0.642    -.2432392    .1500547
            2010  |  -.1994062   .1285302    -1.55   0.121    -.4513208    .0525083
            2011  |   -1.28804   .1263013   -10.20   0.000    -1.535586   -1.040494
            2012  |  -.0760383   .0998479    -0.76   0.446    -.2717367    .1196601
            2014  |  -.3199023    .118299    -2.70   0.007     -.551764   -.0880405
            2015  |  -1.315342   .1105172   -11.90   0.000    -1.531952   -1.098733
            2016  |   .0400198   .0987254     0.41   0.685    -.1534785    .2335181
                  |
    congdist#year |
          2 1999  |   .8181673   .2508889     3.26   0.001     .3264341      1.3099
          2 2000  |   .3399503   .1235665     2.75   0.006     .0977643    .5821362
          2 2003  |   .4430949    .137587     3.22   0.001     .1734294    .7127604
          2 2004  |   .4917483   .1213859     4.05   0.000     .2538364    .7296603
          2 2006  |   .0893556   .1797324     0.50   0.619    -.2629134    .4416246
          2 2007  |   .5303497   .1402421     3.78   0.000     .2554803    .8052191
          2 2008  |   .5410229   .1287261     4.20   0.000     .2887244    .7933213
          2 2010  |   .1233295   .1557374     0.79   0.428    -.1819101    .4285691
          2 2011  |   .2186213   .1459937     1.50   0.134    -.0675211    .5047637
          2 2012  |   .2833521   .1294849     2.19   0.029     .0295663    .5371378
          2 2014  |   .0973128   .1436303     0.68   0.498    -.1841975     .378823
          2 2015  |   .3316126   .1353743     2.45   0.014     .0662839    .5969413
          2 2016  |   .3209086   .1228012     2.61   0.009     .0802227    .5615944
          5 1999  |   .4470272   .3352634     1.33   0.182     -.210077    1.104131
          5 2000  |   .0319493   .2064595     0.15   0.877    -.3727039    .4366026
          5 2003  |   .2403546   .2275492     1.06   0.291    -.2056337     .686343
          5 2004  |   .2193011   .1937012     1.13   0.258    -.1603463    .5989484
          5 2006  |   .0694967   .1956423     0.36   0.722    -.3139552    .4529486
          5 2007  |   .2742606   .2067804     1.33   0.185    -.1310214    .6795427
          5 2008  |   .2511971   .1860835     1.35   0.177    -.1135198     .615914
          5 2010  |   .1972901   .1975107     1.00   0.318    -.1898237    .5844039
          5 2011  |  -.0043405   .2037723    -0.02   0.983    -.4037268    .3950458
          5 2012  |   .2221857   .1521946     1.46   0.144    -.0761103    .5204817
          5 2014  |   .0905075   .1599945     0.57   0.572    -.2230759    .4040908
          5 2015  |   .1675662   .1936916     0.87   0.387    -.2120623    .5471947
          5 2016  |   .1755064   .1514235     1.16   0.246    -.1212781     .472291
          6 1999  |    .636263   .2216921     2.87   0.004     .2017545    1.070772
          6 2000  |   .5069338   .0985842     5.14   0.000     .3137124    .7001553
          6 2003  |   .6638974   .1147594     5.79   0.000     .4389731    .8888218
          6 2004  |   .6210977   .0955025     6.50   0.000     .4339164    .8082791
          6 2006  |   .1139486   .1554851     0.73   0.464    -.1907966    .4186939
          6 2007  |   .6528243   .1187372     5.50   0.000     .4201037     .885545
          6 2008  |   .6252111    .100326     6.23   0.000     .4285759    .8218464
          6 2010  |   .3911783   .1285358     3.04   0.002     .1392527    .6431039
          6 2011  |   .4120025   .1263032     3.26   0.001     .1644528    .6595523
          6 2012  |          0  (empty)
          6 2014  |          0  (empty)
          6 2015  |          0  (empty)
          6 2016  |          0  (empty)
                  |
            _cons |    -.44285   .1006348    -4.40   0.000    -.6400906   -.2456094
-----------------------------------------------------------------------------------

. 
. 
. 
. 
. 
. 
. 
. * Appendix: Table 1 replicated using turnout as a proportion of voting-age population rather than of register
> ed voters

. 
. 
. 
. * Note that "votevap" derives from American Community Survey 5-year data that provides estimates in the form 
> of percentages of the county's total population.  Thus estimated populations are typically not integers.

. 
. 
. 
. import delimited "ky_demos.csv", clear
(16 vars, 3,220 obs)

. 
. collapse timezone congdist expansion (sum) totalvap totalvot, by(year code agegroup)

. 
. gen panelvar = agegroup + 10*code

. 
. gen countyyear = year + 10000*code

. 
. xtset panelvar year
       panel variable:  panelvar (strongly balanced)
        time variable:  year, 1998 to 2016, but with gaps
                delta:  1 unit

. 
. 
. 
. * Generate turnout (as percentage of estimated total population in age-sex category) from raw data

. 
. gen totalpct = 100*totalvot/totalvap
(805 missing values generated)

. 
. 
. 
. * Generate turnout (as proportion of estimated total population in age-sex category) for GLM

. 
. gen totalprop = totalpct/100
(805 missing values generated)

. 
. 
. 
. gen interact = agegroup*timezone

. 
. 
. 
. xtpcse totalpct timezone agegroup interact congdist##year if expansion == 1

Number of gaps in sample:  140

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =        490
Time variable:    year                          Number of groups  =         70
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =          7
                                                              avg =          7
                                                              max =          7
Estimated covariances      =      2485          R-squared         =     0.9025
Estimated autocorrelations =         0          Wald chi2(10)     =    2253.89
Estimated coefficients     =        24          Prob > chi2       =     0.0000

-------------------------------------------------------------------------------
              |           Panel-corrected
     totalpct |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |   4.883737   1.007771     4.85   0.000     2.908542    6.858931
     agegroup |   9.304459   .5162919    18.02   0.000     8.292545    10.31637
     interact |  -1.470965   .1758297    -8.37   0.000    -1.815585   -1.126345
              |
     congdist |
           2  |   1.690885   .7363816     2.30   0.022     .2476036    3.134167
           5  |  -2.714667    .723115    -3.75   0.000    -4.131947   -1.297388
              |
         year |
        2010  |  -6.913655   .6763748   -10.22   0.000    -8.239325   -5.587984
        2011  |   -28.3473   .6763748   -41.91   0.000    -29.67297   -27.02163
        2012  |   -.610471   .8556988    -0.71   0.476     -2.28761    1.066668
        2014  |  -6.658626   .8556988    -7.78   0.000    -8.335764   -4.981487
        2015  |  -25.62319   .8556988   -29.94   0.000    -27.30033   -23.94605
        2016  |   3.109729   1.702636     1.83   0.068    -.2273756    6.446834
              |
congdist#year |
      2 2010  |  -4.097499   1.014562    -4.04   0.000    -6.086005   -2.108994
      2 2011  |  -2.138659   1.014562    -2.11   0.035    -4.127164   -.1501531
      2 2012  |  -5.417835     1.4072    -3.85   0.000    -8.175897   -2.659773
      2 2014  |  -7.471527     1.4072    -5.31   0.000    -10.22959   -4.713465
      2 2015  |  -2.590474     1.4072    -1.84   0.066    -5.348535    .1675877
      2 2016  |  -3.779097   2.292278    -1.65   0.099    -8.271879    .7136844
      5 2010  |   -.405837   .6763748    -0.60   0.548    -1.731507    .9198334
      5 2011  |  -.1818017   .6763748    -0.27   0.788    -1.507472    1.143869
      5 2012  |  -2.640293   .8556988    -3.09   0.002    -4.317431   -.9631538
      5 2014  |  -2.959319   .8556988    -3.46   0.001    -4.636458    -1.28218
      5 2015  |   -.287205   .8556988    -0.34   0.737    -1.964344    1.389934
      5 2016  |  -2.127345   1.702636    -1.25   0.212    -5.464449     1.20976
              |
        _cons |   24.62217   2.064863    11.92   0.000     20.57511    28.66923
-------------------------------------------------------------------------------

. 
. xtpcse totalpct timezone##agegroup congdist##year if expansion == 1

Number of gaps in sample:  140

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =        490
Time variable:    year                          Number of groups  =         70
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =          7
                                                              avg =          7
                                                              max =          7
Estimated covariances      =      2485          R-squared         =     0.9133
Estimated autocorrelations =         0          Wald chi2(13)     =   35583.56
Estimated coefficients     =        30          Prob > chi2       =     0.0000

-----------------------------------------------------------------------------------
                  |           Panel-corrected
         totalpct |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
       1.timezone |   2.265728   .8041589     2.82   0.005     .6896059    3.841851
                  |
         agegroup |
               2  |   7.583085   .6082834    12.47   0.000     6.390872    8.775299
               3  |   18.16246   1.095601    16.58   0.000     16.01512    20.30979
               4  |   30.59279   1.520915    20.11   0.000     27.61186    33.57373
               5  |   35.01744   2.227205    15.72   0.000      30.6522    39.38268
                  |
timezone#agegroup |
             1 2  |     .55548   .3968467     1.40   0.162    -.2223253    1.333285
             1 3  |   -.928726   .5935493    -1.56   0.118    -2.092061    .2346092
             1 4  |  -3.048201   .2939028   -10.37   0.000     -3.62424   -2.472162
             1 5  |  -5.552984   .9339445    -5.95   0.000    -7.383482   -3.722487
                  |
         congdist |
               2  |   1.690885   .7363816     2.30   0.022     .2476036    3.134167
               5  |  -2.714667    .723115    -3.75   0.000    -4.131947   -1.297388
                  |
             year |
            2010  |  -6.913655   .6763748   -10.22   0.000    -8.239325   -5.587984
            2011  |   -28.3473   .6763748   -41.91   0.000    -29.67297   -27.02163
            2012  |   -.610471   .8556988    -0.71   0.476     -2.28761    1.066668
            2014  |  -6.658626   .8556988    -7.78   0.000    -8.335764   -4.981487
            2015  |  -25.62319   .8556988   -29.94   0.000    -27.30033   -23.94605
            2016  |   3.109729   1.702636     1.83   0.068    -.2273756    6.446834
                  |
    congdist#year |
          2 2010  |  -4.097499   1.014562    -4.04   0.000    -6.086005   -2.108994
          2 2011  |  -2.138659   1.014562    -2.11   0.035    -4.127164   -.1501531
          2 2012  |  -5.417835     1.4072    -3.85   0.000    -8.175897   -2.659773
          2 2014  |  -7.471527     1.4072    -5.31   0.000    -10.22959   -4.713465
          2 2015  |  -2.590474     1.4072    -1.84   0.066    -5.348535    .1675877
          2 2016  |  -3.779097   2.292278    -1.65   0.099    -8.271879    .7136844
          5 2010  |   -.405837   .6763748    -0.60   0.548    -1.731507    .9198334
          5 2011  |  -.1818017   .6763748    -0.27   0.788    -1.507472    1.143869
          5 2012  |  -2.640293   .8556988    -3.09   0.002    -4.317431   -.9631538
          5 2014  |  -2.959319   .8556988    -3.46   0.001    -4.636458    -1.28218
          5 2015  |   -.287205   .8556988    -0.34   0.737    -1.964344    1.389934
          5 2016  |  -2.127345   1.702636    -1.25   0.212    -5.464449     1.20976
                  |
            _cons |   34.26439   1.561555    21.94   0.000      31.2038    37.32498
-----------------------------------------------------------------------------------

. 
. glm totalprop timezone agegroup interact congdist##year if expansion == 1, cl(countyyear) family(binomial)
note: totalprop has noninteger values

Iteration 0:   log pseudolikelihood = -202.49079  
Iteration 1:   log pseudolikelihood = -202.31615  
Iteration 2:   log pseudolikelihood = -202.31601  
Iteration 3:   log pseudolikelihood = -202.31601  

Generalized linear models                         No. of obs      =        490
Optimization     : ML                             Residual df     =        466
                                                  Scale parameter =          1
Deviance         =   7.27933128                   (1/df) Deviance =   .0156209
Pearson          =  7.240639089                   (1/df) Pearson  =   .0155379

Variance function: V(u) = u*(1-u/1)               [Binomial]
Link function    : g(u) = ln(u/(1-u))             [Logit]

                                                  AIC             =   .9237388
Log pseudolikelihood = -202.3160057               BIC             =  -2879.314

                             (Std. Err. adjusted for 98 clusters in countyyear)
-------------------------------------------------------------------------------
              |               Robust
    totalprop |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |   .2563823   .0793705     3.23   0.001     .1008189    .4119456
     agegroup |   .4269626   .0174198    24.51   0.000     .3928204    .4611049
     interact |  -.0754995   .0215694    -3.50   0.000    -.1177748   -.0332243
              |
     congdist |
           2  |   .0762733   .0923186     0.83   0.409    -.1046679    .2572144
           5  |  -.1088203   .1723358    -0.63   0.528    -.4465922    .2289516
              |
         year |
        2010  |  -.2959177   .1421465    -2.08   0.037    -.5745197   -.0173156
        2011  |  -1.315986   .0969673   -13.57   0.000    -1.506038   -1.125933
        2012  |  -.0243584   .0849448    -0.29   0.774     -.190847    .1421303
        2014  |  -.2842094   .1116835    -2.54   0.011    -.5031049   -.0653138
        2015  |  -1.169794   .0941721   -12.42   0.000    -1.354368   -.9852205
        2016  |   .1299615   .0825934     1.57   0.116    -.0319186    .2918416
              |
congdist#year |
      2 2010  |  -.1797535    .179788    -1.00   0.317    -.5321315    .1726245
      2 2011  |  -.1085056   .1245178    -0.87   0.384     -.352556    .1355447
      2 2012  |  -.2360312   .1146856    -2.06   0.040    -.4608109   -.0112515
      2 2014  |  -.3288748   .1397886    -2.35   0.019    -.6028554   -.0548942
      2 2015  |  -.1254208   .1265149    -0.99   0.322    -.3733854    .1225438
      2 2016  |    -.15817   .1119724    -1.41   0.158    -.3776318    .0612919
      5 2010  |  -.0254813   .2499229    -0.10   0.919    -.5153212    .4643586
      5 2011  |  -.0861502   .1879075    -0.46   0.647    -.4544421    .2821417
      5 2012  |  -.1175913   .2196003    -0.54   0.592        -.548    .3128174
      5 2014  |  -.1402393   .1919577    -0.73   0.465    -.5164695    .2359908
      5 2015  |  -.0732907   .2223141    -0.33   0.742    -.5090184     .362437
      5 2016  |  -.0871261   .2130941    -0.41   0.683    -.5047829    .3305307
              |
        _cons |  -1.179878   .0807708   -14.61   0.000    -1.338185    -1.02157
-------------------------------------------------------------------------------

. 
. glm totalprop timezone##agegroup congdist##year if expansion == 1, cl(countyyear) family(binomial)
note: totalprop has noninteger values

Iteration 0:   log pseudolikelihood = -202.09828  
Iteration 1:   log pseudolikelihood = -201.90945  
Iteration 2:   log pseudolikelihood = -201.90927  
Iteration 3:   log pseudolikelihood = -201.90927  

Generalized linear models                         No. of obs      =        490
Optimization     : ML                             Residual df     =        460
                                                  Scale parameter =          1
Deviance         =  6.465854032                   (1/df) Deviance =   .0140562
Pearson          =  6.427682704                   (1/df) Pearson  =   .0139732

Variance function: V(u) = u*(1-u/1)               [Binomial]
Link function    : g(u) = ln(u/(1-u))             [Logit]

                                                  AIC             =   .9465684
Log pseudolikelihood =  -201.909267               BIC             =  -2842.961

                                 (Std. Err. adjusted for 98 clusters in countyyear)
-----------------------------------------------------------------------------------
                  |               Robust
        totalprop |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
------------------+----------------------------------------------------------------
       1.timezone |     .13779   .0660729     2.09   0.037     .0082895    .2672904
                  |
         agegroup |
               2  |   .4077236   .0390174    10.45   0.000     .3312509    .4841963
               3  |   .8979742   .0426912    21.03   0.000     .8143009    .9816474
               4  |   1.434265   .0603836    23.75   0.000     1.315915    1.552614
               5  |   1.626935   .0725658    22.42   0.000     1.484709    1.769162
                  |
timezone#agegroup |
             1 2  |   .0001522   .0514708     0.00   0.998    -.1007287    .1010332
             1 3  |  -.0819189    .059983    -1.37   0.172    -.1994835    .0356456
             1 4  |  -.1749166   .0731873    -2.39   0.017     -.318361   -.0314722
             1 5  |  -.2844851   .0948995    -3.00   0.003    -.4704848   -.0984854
                  |
         congdist |
               2  |   .0765013   .0926479     0.83   0.409    -.1050852    .2580878
               5  |  -.1092347   .1728303    -0.63   0.527    -.4479759    .2295064
                  |
             year |
            2010  |  -.2967532   .1425274    -2.08   0.037    -.5761017   -.0174047
            2011  |  -1.317057   .0969767   -13.58   0.000    -1.507128   -1.126986
            2012  |  -.0244578   .0852565    -0.29   0.774    -.1915575    .1426418
            2014  |  -.2850237   .1119911    -2.55   0.011    -.5045223   -.0655252
            2015  |  -1.171011   .0942513   -12.42   0.000     -1.35574   -.9862814
            2016  |   .1304778   .0828999     1.57   0.116    -.0320031    .2929587
                  |
    congdist#year |
          2 2010  |  -.1801903   .1802299    -1.00   0.317    -.5334343    .1730537
          2 2011  |  -.1086222   .1246555    -0.87   0.384    -.3529426    .1356981
          2 2012  |   -.236732   .1150767    -2.06   0.040    -.4622783   -.0111857
          2 2014  |  -.3295319    .140167    -2.35   0.019    -.6042541   -.0548096
          2 2015  |  -.1255867   .1267008    -0.99   0.322    -.3739157    .1227423
          2 2016  |  -.1587932   .1123489    -1.41   0.158     -.378993    .0614065
          5 2010  |  -.0253253   .2505111    -0.10   0.919    -.5163181    .4656675
          5 2011  |  -.0851848   .1882916    -0.45   0.651    -.4542295    .2838598
          5 2012  |  -.1178415   .2202002    -0.54   0.593     -.549426    .3137429
          5 2014  |  -.1402342   .1924937    -0.73   0.466    -.5175149    .2370465
          5 2015  |  -.0724781   .2225714    -0.33   0.745      -.50871    .3637538
          5 2016  |  -.0875227   .2137074    -0.41   0.682    -.5063815     .331336
                  |
            _cons |  -.7745601   .0759997   -10.19   0.000    -.9235168   -.6256034
-----------------------------------------------------------------------------------

. 
. 
. 
. 
. 
. 
. 
. * Appendix: Analysis for placebo test

. 
. 
. 
. * Open raw voting and registration data for placebo test

. 
. import delimited "ky_placebo.csv", clear
(9 vars, 5,880 obs)

. 
. 
. 
. * Label variables (for reference: they will be reprocessed)

. 
. label var year "Year of election"

. 
. label var code "County, numbered alphabetically per Kentucky Secretary of State"

. 
. label var agegroup "Age: 1 = Under 25; 2 = 25-34; 3 = 35-49; 4 = 50-61; 5 = 62 & Over"

. 
. label var female "Sex: 1 = Female; 0 = male"

. 
. label var register "Total number of registrants in age-gender-party-county-year"

. 
. label var voted "Total number of voters in age-gender-party-county-year"

. 
. label var part "Party of registration: d = Democratic, r = Republican, o = Other"

. 
. label var timezone "Time Zone: 0 = Eastern Time; 1 = Central Time"

. 
. label var congdist "Congressional District: 1 = 1st; 2 = 2nd; 5 = 5th"

. 
. 
. 
. * Aggregate data by gender 

. 
. collapse timezone congdist (sum) register voted, by(year code agegroup part)

. 
. 
. 
. * Create numeric party codes

. 
. encode part, gen(partyno)

. 
. 
. 
. * Create variable marking age-county-party to panel-correct standard errors

. 
. gen panelvar = partyno + 10*agegroup + 100*code

. 
. 
. 
. * Generate turnout as percentage of registered voters

. 
. gen turnout = 100*voted/register

. 
. 
. 
. * Label variables

. 
. label var timezone "Time Zone: 0 = Eastern Time; 1 = Central Time"

. 
. label var congdist "Congressional District: 1 = 1st; 2 = 2nd; 5 = 5th"

. 
. label var panelvar "Code for party (last digit), age (penultimate digit), county (earlier digits)"

. 
. label var register "Total number of registrants in age-party-county-year"

. 
. label var voted "Total number of voters in age-party-county-year"

. 
. label var turnout "Percent of age-county-party registered voting in election year"

. 
. 
. 
. * Set up panel by county-age-party (cross-section) and year (time-series)

. 
. xtset panelvar year
       panel variable:  panelvar (strongly balanced)
        time variable:  year, 1998 to 2016, but with gaps
                delta:  1 unit

. 
. 
. 
. * Estimate placebo-test model

. 
. xtpcse turnout timezone##agegroup##partyno congdist##year

Number of gaps in sample:  840

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   panelvar                      Number of obs     =      2,940
Time variable:    year                          Number of groups  =        210
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =         14
                                                              avg =         14
                                                              max =         14
Estimated covariances      =     22155          R-squared         =     0.8922
Estimated autocorrelations =         0          Wald chi2(27)     =    6165.71
Estimated coefficients     =        71          Prob > chi2       =     0.0000

-------------------------------------------------------------------------------------------
                          |           Panel-corrected
                  turnout |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------------------+----------------------------------------------------------------
               1.timezone |   2.464805   1.060676     2.32   0.020     .3859179    4.543691
                          |
                 agegroup |
                       2  |   5.201835   .8861147     5.87   0.000     3.465082    6.938588
                       3  |   17.99508   1.322508    13.61   0.000     15.40301    20.58714
                       4  |   29.28552   1.676879    17.46   0.000      25.9989    32.57214
                       5  |   30.54752   2.004437    15.24   0.000      26.6189    34.47614
                          |
        timezone#agegroup |
                     1 2  |  -1.160555   .3817397    -3.04   0.002    -1.908751   -.4123592
                     1 3  |  -1.678317   .4484582    -3.74   0.000    -2.557278   -.7993546
                     1 4  |  -3.408449    .489202    -6.97   0.000    -4.367267   -2.449631
                     1 5  |   -4.46826    .506507    -8.82   0.000    -5.460995   -3.475524
                          |
                  partyno |
                       o  |  -9.378095    .981787    -9.55   0.000    -11.30236   -7.453827
                       r  |   3.689453   .7982648     4.62   0.000     2.124882    5.254023
                          |
         timezone#partyno |
                     1#o  |   -1.94168   1.017293    -1.91   0.056    -3.935539    .0521782
                     1#r  |   .4011076   .3640109     1.10   0.271    -.3123407    1.114556
                          |
         agegroup#partyno |
                     2#o  |  -4.620525   .8466731    -5.46   0.000    -6.279974   -2.961077
                     2#r  |   .3597686    .388761     0.93   0.355     -.402189    1.121726
                     3#o  |  -8.404056   .8999009    -9.34   0.000    -10.16783   -6.640282
                     3#r  |  -.7573782   .4949978    -1.53   0.126    -1.727556    .2127997
                     4#o  |  -9.141427   1.145591    -7.98   0.000    -11.38674   -6.896109
                     4#r  |  -1.729872   .4232099    -4.09   0.000    -2.559348   -.9003959
                     5#o  |  -3.797178   1.399063    -2.71   0.007     -6.53929   -1.055065
                     5#r  |  -1.222714    .449122    -2.72   0.006    -2.102977   -.3424513
                          |
timezone#agegroup#partyno |
                   1#2#o  |   .6935646   1.144107     0.61   0.544    -1.548845    2.935974
                   1#2#r  |  -.0382289   .6399733    -0.06   0.952    -1.292553    1.216096
                   1#3#o  |  -.6906908   1.093234    -0.63   0.528    -2.833391    1.452009
                   1#3#r  |  -.3900634   .3334915    -1.17   0.242    -1.043695    .2635679
                   1#4#o  |   .6268999   1.032818     0.61   0.544    -1.397386    2.651186
                   1#4#r  |  -.8046848   .4462538    -1.80   0.071    -1.679326    .0699566
                   1#5#o  |  -.1820979   .9089095    -0.20   0.841    -1.963528    1.599332
                   1#5#r  |   -2.34137   .5581897    -4.19   0.000    -3.435401   -1.247338
                          |
                 congdist |
                       2  |  -7.170006   1.917053    -3.74   0.000    -10.92736   -3.412651
                       5  |  -4.430245   1.564199    -2.83   0.005    -7.496017   -1.364472
                          |
                     year |
                    1999  |  -36.26818   2.018388   -17.97   0.000    -40.22415   -32.31221
                    2000  |  -.5499147   2.018388    -0.27   0.785    -4.505883    3.406054
                    2003  |   -18.7643   1.542092   -12.17   0.000    -21.78674   -15.74185
                    2004  |    3.53092   1.542092     2.29   0.022     .5084749    6.553365
                    2006  |   -1.71571   1.542092    -1.11   0.266    -4.738155    1.306734
                    2007  |  -20.54301   1.542092   -13.32   0.000    -23.56546   -17.52057
                    2008  |   2.417953   1.542092     1.57   0.117     -.604492    5.440398
                    2010  |  -5.374107   1.542092    -3.48   0.000    -8.396552   -2.351662
                    2011  |  -25.44212   1.542092   -16.50   0.000    -28.46457   -22.41968
                    2012  |   1.335997   1.634259     0.82   0.414    -1.867092    4.539087
                    2014  |  -7.115115   1.634259    -4.35   0.000     -10.3182   -3.912025
                    2015  |  -24.98566   1.634259   -15.29   0.000    -28.18875   -21.78257
                    2016  |   4.116603   2.018388     2.04   0.041      .160634    8.072571
                          |
            congdist#year |
                  2 1999  |   13.95001   2.691185     5.18   0.000     8.675381    19.22463
                  2 2000  |   7.074064   2.691185     2.63   0.009     1.799439    12.34869
                  2 2003  |   7.962306   2.094979     3.80   0.000     3.856221    12.06839
                  2 2004  |   10.23114   2.094979     4.88   0.000     6.125051    14.33722
                  2 2006  |   3.775192   2.094979     1.80   0.072    -.3308922    7.881276
                  2 2007  |   7.948115   2.094979     3.79   0.000     3.842031     12.0542
                  2 2008  |   10.68127   2.094979     5.10   0.000     6.575185    14.78735
                  2 2010  |   5.992681   2.094979     2.86   0.004     1.886597    10.09876
                  2 2011  |   6.174399   2.094979     2.95   0.003     2.068315    10.28048
                  2 2012  |   5.418446   2.333085     2.32   0.020     .8456834    9.991208
                  2 2014  |   3.213006   2.333085     1.38   0.168    -1.359757    7.785768
                  2 2015  |   7.262363   2.333085     3.11   0.002       2.6896    11.83513
                  2 2016  |   6.117338   2.691185     2.27   0.023      .842713    11.39196
                  5 1999  |   8.280839   2.018388     4.10   0.000     4.324871    12.23681
                  5 2000  |   1.583654   2.018388     0.78   0.433    -2.372314    5.539623
                  5 2003  |   3.453926   1.542092     2.24   0.025     .4314814    6.476371
                  5 2004  |   3.203842   1.542092     2.08   0.038     .1813971    6.226287
                  5 2006  |  -2.171006   1.542092    -1.41   0.159    -5.193451    .8514385
                  5 2007  |   3.984217   1.542092     2.58   0.010     .9617725    7.006662
                  5 2008  |   3.106915   1.542092     2.01   0.044     .0844702     6.12936
                  5 2010  |   1.935172   1.542092     1.25   0.210    -1.087273    4.957617
                  5 2011  |   1.191588   1.542092     0.77   0.440    -1.830857    4.214033
                  5 2012  |   .3421227   1.634259     0.21   0.834    -2.860967    3.545212
                  5 2014  |  -.3203293   1.634259    -0.20   0.845    -3.523419     2.88276
                  5 2015  |   1.471127   1.634259     0.90   0.368    -1.731962    4.674216
                  5 2016  |   .0524435   2.018388     0.03   0.979    -3.903525    4.008412
                          |
                    _cons |   36.12555   1.980761    18.24   0.000     32.24333    40.00777
-------------------------------------------------------------------------------------------

. 
. 
. 
. * Generate data for figure (should correspond to the data in figappdata.csv)

. 
. margins timezone#agegroup#partyno, pwcompare

Pairwise comparisons of predictive margins
Model VCE    : Panel-corrected

Expression   : Fitted values, predict()

---------------------------------------------------------------------------
                          |            Delta-method         Unadjusted
                          |   Contrast   Std. Err.     [95% Conf. Interval]
--------------------------+------------------------------------------------
timezone#agegroup#partyno |
      (0#1#o) vs (0#1#d)  |  -9.378095    .981787     -11.30236   -7.453827
      (0#1#r) vs (0#1#d)  |   3.689453   .7982648      2.124882    5.254023
      (0#2#d) vs (0#1#d)  |   5.201835   .8861147      3.465082    6.938588
      (0#2#o) vs (0#1#d)  |  -8.796785    .996969     -10.75081   -6.842761
      (0#2#r) vs (0#1#d)  |   9.251057   .8762385      7.533661    10.96845
      (0#3#d) vs (0#1#d)  |   17.99508   1.322508      15.40301    20.58714
      (0#3#o) vs (0#1#d)  |   .2129252   .7813274     -1.318448    1.744299
      (0#3#r) vs (0#1#d)  |   20.92715   1.296224       18.3866     23.4677
      (0#4#d) vs (0#1#d)  |   29.28552   1.676879       25.9989    32.57214
      (0#4#o) vs (0#1#d)  |     10.766   1.069612      8.669597     12.8624
      (0#4#r) vs (0#1#d)  |    31.2451   1.648978      28.01316    34.47704
      (0#5#d) vs (0#1#d)  |   30.54752   2.004437       26.6189    34.47614
      (0#5#o) vs (0#1#d)  |   17.37225    1.74552      13.95109     20.7934
      (0#5#r) vs (0#1#d)  |   33.01426   2.124635      28.85005    37.17847
      (1#1#d) vs (0#1#d)  |   2.464805   1.060676      .3859179    4.543691
      (1#1#o) vs (0#1#d)  |   -8.85497   1.157771     -11.12416   -6.585782
      (1#1#r) vs (0#1#d)  |   6.555365   1.194624      4.213945    8.896785
      (1#2#d) vs (0#1#d)  |   6.506085   1.728244      3.118788    9.893381
      (1#2#o) vs (0#1#d)  |  -8.740651    1.45735     -11.59701   -5.884296
      (1#2#r) vs (0#1#d)  |   10.91818   1.604096      7.774214    14.06216
      (1#3#d) vs (0#1#d)  |   18.78156   2.069139      14.72613      22.837
      (1#3#o) vs (0#1#d)  |  -1.632958   1.436085     -4.447633    1.181718
      (1#3#r) vs (0#1#d)  |   21.72468   2.084185      17.63976    25.80961
      (1#4#d) vs (0#1#d)  |   28.34188   2.180804      24.06758    32.61617
      (1#4#o) vs (0#1#d)  |   8.507574   1.534012      5.500967    11.51418
      (1#4#r) vs (0#1#d)  |   29.89788   2.352605      25.28686     34.5089
      (1#5#d) vs (0#1#d)  |   28.54406   2.489261       23.6652    33.42293
      (1#5#o) vs (0#1#d)  |   13.24501   1.629057      10.05212    16.43791
      (1#5#r) vs (0#1#d)  |   29.07054   2.693757      23.79087    34.35021
      (0#1#r) vs (0#1#o)  |   13.06755   1.186356      10.74233    15.39276
      (0#2#d) vs (0#1#o)  |   14.57993   1.441115       11.7554    17.40446
      (0#2#o) vs (0#1#o)  |     .58131   .9418891     -1.264759    2.427379
      (0#2#r) vs (0#1#o)  |   18.62915   1.396481       15.8921     21.3662
      (0#3#d) vs (0#1#o)  |   27.37317   1.915848      23.61818    31.12816
      (0#3#o) vs (0#1#o)  |    9.59102   1.076049      7.482003    11.70004
      (0#3#r) vs (0#1#o)  |   30.30524   1.746133      26.88289     33.7276
      (0#4#d) vs (0#1#o)  |   38.66361   2.111391      34.52536    42.80187
      (0#4#o) vs (0#1#o)  |   20.14409   1.545918      17.11415    23.17404
      (0#4#r) vs (0#1#o)  |    40.6232   1.986331      36.73006    44.51633
      (0#5#d) vs (0#1#o)  |   39.92561   2.069925      35.86864    43.98259
      (0#5#o) vs (0#1#o)  |   26.75034   1.814178      23.19462    30.30606
      (0#5#r) vs (0#1#o)  |   42.39235   2.179352       38.1209     46.6638
      (1#1#d) vs (0#1#o)  |    11.8429   1.585978      8.734439    14.95136
      (1#1#o) vs (0#1#o)  |   .5231244   .8023043     -1.049363    2.095612
      (1#1#r) vs (0#1#o)  |   15.93346   1.675978       12.6486    19.21832
      (1#2#d) vs (0#1#o)  |   15.88418   2.213686      11.54544    20.22292
      (1#2#o) vs (0#1#o)  |   .6374438   1.287161     -1.885345    3.160232
      (1#2#r) vs (0#1#o)  |   20.29628   2.002963      16.37054    24.22201
      (1#3#d) vs (0#1#o)  |   28.15966   2.532439      23.19617    33.12315
      (1#3#o) vs (0#1#o)  |   7.745137   1.559685      4.688211    10.80206
      (1#3#r) vs (0#1#o)  |   31.10278   2.478756       26.2445    35.96105
      (1#4#d) vs (0#1#o)  |   37.71997   2.594366      32.63511    42.80483
      (1#4#o) vs (0#1#o)  |   17.88567   1.856704       14.2466    21.52474
      (1#4#r) vs (0#1#o)  |   39.27597    2.71421      33.95622    44.59573
      (1#5#d) vs (0#1#o)  |   37.92216   2.605973      32.81455    43.02977
      (1#5#o) vs (0#1#o)  |   22.62311   1.843322      19.01026    26.23595
      (1#5#r) vs (0#1#o)  |   38.44864   2.769537      33.02044    43.87683
      (0#2#d) vs (0#1#r)  |   1.512383   1.427548     -1.285559    4.310325
      (0#2#o) vs (0#1#r)  |  -12.48624   1.415057      -15.2597   -9.712776
      (0#2#r) vs (0#1#r)  |   5.561604   .8499235      3.895785    7.227423
      (0#3#d) vs (0#1#r)  |   14.30562   1.742841      10.88972    17.72153
      (0#3#o) vs (0#1#r)  |  -3.476527   1.125416     -5.682302   -1.270753
      (0#3#r) vs (0#1#r)  |    17.2377   1.242319       14.8028     19.6726
      (0#4#d) vs (0#1#r)  |   25.59607   2.033052      21.61136    29.58078
      (0#4#o) vs (0#1#r)  |   7.076546   1.501133      4.134379    10.01871
      (0#4#r) vs (0#1#r)  |   27.55565    1.65644      24.30908    30.80221
      (0#5#d) vs (0#1#r)  |   26.85807   2.268353      22.41218    31.30396
      (0#5#o) vs (0#1#r)  |   13.68279    1.72405      10.30372    17.06187
      (0#5#r) vs (0#1#r)  |   29.32481   2.114375      25.18071     33.4689
      (1#1#d) vs (0#1#r)  |  -1.224648   1.454503     -4.075422    1.626126
      (1#1#o) vs (0#1#r)  |  -12.54442   1.373291     -15.23602   -9.852823
      (1#1#r) vs (0#1#r)  |   2.865912   1.011955      .8825175    4.849307
      (1#2#d) vs (0#1#r)  |   2.816632   2.174148      -1.44462    7.077884
      (1#2#o) vs (0#1#r)  |   -12.4301   1.769206     -15.89768   -8.962523
      (1#2#r) vs (0#1#r)  |   7.228732   1.636614      4.021028    10.43644
      (1#3#d) vs (0#1#r)  |   15.09211   2.379024      10.42931    19.75491
      (1#3#o) vs (0#1#r)  |   -5.32241   1.746838     -8.746149   -1.898672
      (1#3#r) vs (0#1#r)  |   18.03523   2.059571      13.99854    22.07191
      (1#4#d) vs (0#1#r)  |   24.65242   2.491382       19.7694    29.53544
      (1#4#o) vs (0#1#r)  |   4.818121   1.783754      1.322027    8.314216
      (1#4#r) vs (0#1#r)  |   26.20843   2.356569      21.58964    30.82722
      (1#5#d) vs (0#1#r)  |   24.85461   2.711903      19.53938    30.16984
      (1#5#o) vs (0#1#r)  |   9.555561   1.741482       6.14232     12.9688
      (1#5#r) vs (0#1#r)  |   25.38109    2.66919      20.14957     30.6126
      (0#2#o) vs (0#2#d)  |  -13.99862   1.073012     -16.10169   -11.89555
      (0#2#r) vs (0#2#d)  |   4.049221   .9011247      2.283049    5.815393
      (0#3#d) vs (0#2#d)  |   12.79324   .5401567      11.73455    13.85193
      (0#3#o) vs (0#2#d)  |   -4.98891   .6831122     -6.327786   -3.650035
      (0#3#r) vs (0#2#d)  |   15.72531   1.034206      13.69831    17.75232
      (0#4#d) vs (0#2#d)  |   24.08368   .8749339      22.36885    25.79852
      (0#4#o) vs (0#2#d)  |   5.564163   .4967781      4.590496    6.537831
      (0#4#r) vs (0#2#d)  |   26.04327    1.22868       23.6351    28.45143
      (0#5#d) vs (0#2#d)  |   25.34568    1.35603      22.68791    28.00345
      (0#5#o) vs (0#2#d)  |   12.17041   1.425856      9.375786    14.96504
      (0#5#r) vs (0#2#d)  |   27.81242   1.704867      24.47094     31.1539
      (1#1#d) vs (0#2#d)  |  -2.737031   .7375798     -4.182661   -1.291401
      (1#1#o) vs (0#2#d)  |  -14.05681   1.266971     -16.54002   -11.57359
      (1#1#r) vs (0#2#d)  |    1.35353   1.274143     -1.143744    3.850803
      (1#2#d) vs (0#2#d)  |   1.304249   .9627081     -.5826239    3.191123
      (1#2#o) vs (0#2#d)  |  -13.94249   1.193678     -16.28205   -11.60292
      (1#2#r) vs (0#2#d)  |   5.716349   1.188665      3.386609     8.04609
      (1#3#d) vs (0#2#d)  |   13.57973    1.30202      11.02782    16.13164
      (1#3#o) vs (0#2#d)  |  -6.834793    .848238     -8.497309   -5.172277
      (1#3#r) vs (0#2#d)  |   16.52285   1.606286      13.37459    19.67111
      (1#4#d) vs (0#2#d)  |   23.14004   1.404751      20.38678     25.8933
      (1#4#o) vs (0#2#d)  |   3.305739   .9249168      1.492935    5.118542
      (1#4#r) vs (0#2#d)  |   24.69604   1.809799       21.1489    28.24318
      (1#5#d) vs (0#2#d)  |   23.34223   1.773773       19.8657    26.81876
      (1#5#o) vs (0#2#d)  |   8.043178   1.155077       5.77927    10.30709
      (1#5#r) vs (0#2#d)  |   23.86871   2.165877      19.62366    28.11375
      (0#2#r) vs (0#2#o)  |   18.04784   1.184183      15.72689     20.3688
      (0#3#d) vs (0#2#o)  |   26.79186   1.545876        23.762    29.82172
      (0#3#o) vs (0#2#o)  |    9.00971   .7228055      7.593037    10.42638
      (0#3#r) vs (0#2#o)  |   29.72393   1.512722      26.75905    32.68882
      (0#4#d) vs (0#2#o)  |    38.0823   1.637854      34.87217    41.29244
      (0#4#o) vs (0#2#o)  |   19.56278   1.073262      17.45923    21.66634
      (0#4#r) vs (0#2#o)  |   40.04189    1.65598      36.79622    43.28755
      (0#5#d) vs (0#2#o)  |    39.3443   1.533447       36.3388     42.3498
      (0#5#o) vs (0#2#o)  |   26.16903   1.443868       23.3391    28.99896
      (0#5#r) vs (0#2#o)  |   41.81104   1.756644      38.36808      45.254
      (1#1#d) vs (0#2#o)  |   11.26159   1.565094      8.194062    14.32912
      (1#1#o) vs (0#2#o)  |  -.0581856   .8410339     -1.706582    1.590211
      (1#1#r) vs (0#2#o)  |   15.35215    1.80259      11.81914    18.88516
      (1#2#d) vs (0#2#o)  |   15.30287   1.943979      11.49274      19.113
      (1#2#o) vs (0#2#o)  |   .0561338   .9531615     -1.812028    1.924296
      (1#2#r) vs (0#2#o)  |   19.71497    1.88656      16.01738    23.41256
      (1#3#d) vs (0#2#o)  |   27.57835   2.226408      23.21467    31.94203
      (1#3#o) vs (0#2#o)  |   7.163827   1.265242      4.683999    9.643655
      (1#3#r) vs (0#2#o)  |   30.52147   2.298305      26.01687    35.02606
      (1#4#d) vs (0#2#o)  |   37.13866   2.259376      32.71036    41.56696
      (1#4#o) vs (0#2#o)  |   17.30436   1.385707      14.58842    20.02029
      (1#4#r) vs (0#2#o)  |   38.69466   2.481254       33.8315    43.55783
      (1#5#d) vs (0#2#o)  |   37.34085   2.182763      33.06271    41.61899
      (1#5#o) vs (0#2#o)  |    22.0418   1.321174      19.45234    24.63125
      (1#5#r) vs (0#2#o)  |   37.86733   2.481669      33.00334    42.73131
      (0#3#d) vs (0#2#r)  |   8.744019   1.109643      6.569158    10.91888
      (0#3#o) vs (0#2#r)  |  -9.038131   .7508533     -10.50978   -7.566486
      (0#3#r) vs (0#2#r)  |   11.67609   .5418738      10.61404    12.73815
      (0#4#d) vs (0#2#r)  |   20.03446   1.275654      17.53423     22.5347
      (0#4#o) vs (0#2#r)  |   1.514942   .9657697     -.3779317    3.407816
      (0#4#r) vs (0#2#r)  |   21.99404   .8965282      20.23688    23.75121
      (0#5#d) vs (0#2#r)  |   21.29646   1.570769      18.21781    24.37511
      (0#5#o) vs (0#2#r)  |    8.12119   1.222658      5.724825    10.51756
      (0#5#r) vs (0#2#r)  |    23.7632   1.444935      20.93118    26.59522
      (1#1#d) vs (0#2#r)  |  -6.786252   1.137255     -9.015231   -4.557273
      (1#1#o) vs (0#2#r)  |  -18.10603   1.336393     -20.72531   -15.48674
      (1#1#r) vs (0#2#r)  |  -2.695692   .8653415      -4.39173   -.9996536
      (1#2#d) vs (0#2#r)  |  -2.744972    1.59071     -5.862706    .3727625
      (1#2#o) vs (0#2#r)  |  -17.99171   1.359864     -20.65699   -15.32642
      (1#2#r) vs (0#2#r)  |   1.667128   1.066751     -.4236653    3.757922
      (1#3#d) vs (0#2#r)  |   9.530507   1.719033      6.161264    12.89975
      (1#3#o) vs (0#2#r)  |  -10.88401   1.246335     -13.32679   -8.441243
      (1#3#r) vs (0#2#r)  |   12.47363   1.443236      9.644935    15.30232
      (1#4#d) vs (0#2#r)  |   19.09082   1.770858         15.62    22.56164
      (1#4#o) vs (0#2#r)  |  -.7434826   1.078541     -2.857384    1.370419
      (1#4#r) vs (0#2#r)  |   20.64682   1.665379      17.38274    23.91091
      (1#5#d) vs (0#2#r)  |   19.29301   1.981244      15.40984    23.17617
      (1#5#o) vs (0#2#r)  |   3.993957   1.076204      1.884636    6.103278
      (1#5#r) vs (0#2#r)  |   19.81948    2.01556      15.86906    23.76991
      (0#3#o) vs (0#3#d)  |  -17.78215   1.104515     -19.94696   -15.61734
      (0#3#r) vs (0#3#d)  |   2.932075   1.016417       .939934    4.924215
      (0#4#d) vs (0#3#d)  |   11.29044   .5571784      10.19839    12.38249
      (0#4#o) vs (0#3#d)  |  -7.229077   .6924917     -8.586336   -5.871818
      (0#4#r) vs (0#3#d)  |   13.25002   1.124396      11.04625     15.4538
      (0#5#d) vs (0#3#d)  |   12.55244   1.325691      9.954136    15.15075
      (0#5#o) vs (0#3#d)  |  -.6228289   1.526727      -3.61516    2.369502
      (0#5#r) vs (0#3#d)  |   15.01918   1.701301      11.68469    18.35367
      (1#1#d) vs (0#3#d)  |  -15.53027   .8034685     -17.10504    -13.9555
      (1#1#o) vs (0#3#d)  |  -26.85005   1.666829     -30.11697   -23.58312
      (1#1#r) vs (0#3#d)  |  -11.43971   1.332931     -14.05221   -8.827215
      (1#2#d) vs (0#3#d)  |  -11.48899    .658111     -12.77886   -10.19912
      (1#2#o) vs (0#3#d)  |  -26.73573   1.499791     -29.67526   -23.79619
      (1#2#r) vs (0#3#d)  |  -7.076891   1.035818     -9.107057   -5.046724
      (1#3#d) vs (0#3#d)  |   .7864881   .8660936     -.9110243       2.484
      (1#3#o) vs (0#3#d)  |  -19.62803    .966673     -21.52268   -17.73339
      (1#3#r) vs (0#3#d)  |   3.729607    1.33338      1.116229    6.342984
      (1#4#d) vs (0#3#d)  |    10.3468   .9872257      8.411873    12.28173
      (1#4#o) vs (0#3#d)  |  -9.487502   .9411767     -11.33217   -7.642829
      (1#4#r) vs (0#3#d)  |    11.9028   1.501171      8.960562    14.84504
      (1#5#d) vs (0#3#d)  |   10.54899    1.57714      7.457852    13.64013
      (1#5#o) vs (0#3#d)  |  -4.750062   1.183764     -7.070198   -2.429927
      (1#5#r) vs (0#3#d)  |   11.07547   2.009918      7.136098    15.01483
      (0#3#r) vs (0#3#o)  |   20.71423   1.090143      18.57758    22.85087
      (0#4#d) vs (0#3#o)  |   29.07259   1.292517      26.53931    31.60588
      (0#4#o) vs (0#3#o)  |   10.55307   .8657093      8.856314    12.24983
      (0#4#r) vs (0#3#o)  |   31.03218   1.308499      28.46756    33.59679
      (0#5#d) vs (0#3#o)  |   30.33459   1.458822      27.47535    33.19383
      (0#5#o) vs (0#3#o)  |   17.15932   1.333846      14.54503    19.77361
      (0#5#r) vs (0#3#o)  |   32.80133   1.617007      29.63206    35.97061
      (1#1#d) vs (0#3#o)  |   2.251879   1.145729      .0062911    4.497468
      (1#1#o) vs (0#3#o)  |  -9.067895   .8565381     -10.74668   -7.389112
      (1#1#r) vs (0#3#o)  |    6.34244   1.329255      3.737147    8.947732
      (1#2#d) vs (0#3#o)  |    6.29316   1.547296      3.260515    9.325804
      (1#2#o) vs (0#3#o)  |  -8.953576    1.02565     -10.96381   -6.943339
      (1#2#r) vs (0#3#o)  |   10.70526   1.423507      7.915238    13.49528
      (1#3#d) vs (0#3#o)  |   18.56864   1.794993      15.05052    22.08676
      (1#3#o) vs (0#3#o)  |  -1.845883   .9478404     -3.703616    .0118501
      (1#3#r) vs (0#3#o)  |   21.51176   1.853043      17.87986    25.14366
      (1#4#d) vs (0#3#o)  |   28.12895   1.874657      24.45469    31.80321
      (1#4#o) vs (0#3#o)  |   8.294649   .9981075      6.338394     10.2509
      (1#4#r) vs (0#3#o)  |   29.68495   2.055831       25.6556    33.71431
      (1#5#d) vs (0#3#o)  |   28.33114   2.002559       24.4062    32.25608
      (1#5#o) vs (0#3#o)  |   13.03209   1.071145      10.93268    15.13149
      (1#5#r) vs (0#3#o)  |   28.85762   2.256399      24.43515    33.28008
      (0#4#d) vs (0#3#r)  |    8.35837   1.061437      6.277992    10.43875
      (0#4#o) vs (0#3#r)  |  -10.16115    1.02034     -12.16098   -8.161322
      (0#4#r) vs (0#3#r)  |   10.31795   .5164222      9.305781    11.33012
      (0#5#d) vs (0#3#r)  |   9.620369   1.363875      6.947223    12.29352
      (0#5#o) vs (0#3#r)  |  -3.554903   1.041718     -5.596633   -1.513174
      (0#5#r) vs (0#3#r)  |   12.08711   1.180257      9.773847    14.40037
      (1#1#d) vs (0#3#r)  |  -18.46235   1.137512     -20.69183   -16.23286
      (1#1#o) vs (0#3#r)  |  -29.78212   1.567136     -32.85365   -26.71059
      (1#1#r) vs (0#3#r)  |  -14.37179   .8166395     -15.97237    -12.7712
      (1#2#d) vs (0#3#r)  |  -14.42107   1.427029     -17.21799   -11.62414
      (1#2#o) vs (0#3#r)  |   -29.6678    1.42035     -32.45164   -26.88397
      (1#2#r) vs (0#3#r)  |  -10.00897   .6780031     -11.33783   -8.680104
      (1#3#d) vs (0#3#r)  |  -2.145586   1.403232      -4.89587    .6046969
      (1#3#o) vs (0#3#r)  |  -22.56011   1.172719      -24.8586   -20.26162
      (1#3#r) vs (0#3#r)  |   .7975322   .9540122     -1.072297    2.667362
      (1#4#d) vs (0#3#r)  |   7.414725   1.438225      4.595856    10.23359
      (1#4#o) vs (0#3#r)  |  -12.41958   1.012251     -14.40355    -10.4356
      (1#4#r) vs (0#3#r)  |   8.970729   1.196846      6.624953     11.3165
      (1#5#d) vs (0#3#r)  |   7.616914   1.621358      4.439111    10.79472
      (1#5#o) vs (0#3#r)  |  -7.682137   .9087567     -9.463267   -5.901006
      (1#5#r) vs (0#3#r)  |   8.143391   1.585683      5.035509    11.25127
      (0#4#o) vs (0#4#d)  |  -18.51952   .8652922     -20.21546   -16.82358
      (0#4#r) vs (0#4#d)  |   1.959581    .915748      .1647476    3.754414
      (0#5#d) vs (0#4#d)  |   1.261999   .9043388     -.5104721    3.034471
      (0#5#o) vs (0#4#d)  |  -11.91327   1.352161     -14.56346   -9.263087
      (0#5#r) vs (0#4#d)  |   3.728738   1.331096      1.119837    6.337638
      (1#1#d) vs (0#4#d)  |  -26.82072   1.216827     -29.20565   -24.43578
      (1#1#o) vs (0#4#d)  |  -38.14049   1.800227     -41.66887   -34.61211
      (1#1#r) vs (0#4#d)  |  -22.73016   1.604231     -25.87439   -19.58592
      (1#2#d) vs (0#4#d)  |  -22.77944   .8652131     -24.47522   -21.08365
      (1#2#o) vs (0#4#d)  |  -38.02617   1.476448     -40.91995   -35.13239
      (1#2#r) vs (0#4#d)  |  -18.36734   1.089176     -20.50208   -16.23259
      (1#3#d) vs (0#4#d)  |  -10.50396   .8087947     -12.08916   -8.918748
      (1#3#o) vs (0#4#d)  |  -30.91848   1.017665     -32.91306   -28.92389
      (1#3#r) vs (0#4#d)  |  -7.560838    1.22607      -9.96389   -5.157785
      (1#4#d) vs (0#4#d)  |  -.9436443   .7336447     -2.381562    .4942729
      (1#4#o) vs (0#4#d)  |  -20.77795   .7786614     -22.30409    -19.2518
      (1#4#r) vs (0#4#d)  |   .6123591   1.256952     -1.851221    3.075939
      (1#5#d) vs (0#4#d)  |  -.7414556   1.121959     -2.940455    1.457544
      (1#5#o) vs (0#4#d)  |  -16.04051   1.006376     -18.01297   -14.06805
      (1#5#r) vs (0#4#d)  |  -.2149791   1.619552     -3.389242    2.959284
      (0#4#r) vs (0#4#o)  |    20.4791   1.139371      18.24598    22.71223
      (0#5#d) vs (0#4#o)  |   19.78152   1.199506      17.43053    22.13251
      (0#5#o) vs (0#4#o)  |   6.606248   1.228473      4.198485    9.014012
      (0#5#r) vs (0#4#o)  |   22.24826   1.553352      19.20375    25.29277
      (1#1#d) vs (0#4#o)  |  -8.301194   1.021587     -10.30347    -6.29892
      (1#1#o) vs (0#4#o)  |  -19.62097   1.309445     -22.18743    -17.0545
      (1#1#r) vs (0#4#o)  |  -4.210634   1.424567     -7.002733   -1.418534
      (1#2#d) vs (0#4#o)  |  -4.259914   1.172181     -6.557346   -1.962482
      (1#2#o) vs (0#4#o)  |  -19.50665   1.305592     -22.06556   -16.94774
      (1#2#r) vs (0#4#o)  |   .1521861   1.284852     -2.366077     2.67045
      (1#3#d) vs (0#4#o)  |   8.015565   1.399194      5.273194    10.75794
      (1#3#o) vs (0#4#o)  |  -12.39896   .9269085     -14.21566   -10.58225
      (1#3#r) vs (0#4#o)  |   10.95868   1.631395      7.761209    14.15616
      (1#4#d) vs (0#4#o)  |   17.57588   1.476038       14.6829    20.46886
      (1#4#o) vs (0#4#o)  |  -2.258425   .9932563     -4.205171   -.3116781
      (1#4#r) vs (0#4#o)  |   19.13188   1.814835      15.57487    22.68889
      (1#5#d) vs (0#4#o)  |   17.77807    1.70563      14.43509    21.12104
      (1#5#o) vs (0#4#o)  |   2.479015   1.068333      .3851206    4.572909
      (1#5#r) vs (0#4#o)  |   18.30454   2.062928      14.26128    22.34781
      (0#5#d) vs (0#4#r)  |  -.6975812   1.088747     -2.831487    1.436325
      (0#5#o) vs (0#4#r)  |  -13.87285   1.036472      -15.9043   -11.84141
      (0#5#r) vs (0#4#r)  |   1.769157   .8236768      .1547804    3.383534
      (1#1#d) vs (0#4#r)  |   -28.7803   1.444992     -31.61243   -25.94816
      (1#1#o) vs (0#4#r)  |  -40.10007   1.742553     -43.51541   -36.68473
      (1#1#r) vs (0#4#r)  |  -24.68974   1.247202     -27.13421   -22.24526
      (1#2#d) vs (0#4#r)  |  -24.73902   1.491164     -27.66164   -21.81639
      (1#2#o) vs (0#4#r)  |  -39.98575   1.510782     -42.94683   -37.02467
      (1#2#r) vs (0#4#r)  |  -20.32692   .8614135     -22.01526   -18.63858
      (1#3#d) vs (0#4#r)  |  -12.46354    1.37477     -15.15804   -9.769037
      (1#3#o) vs (0#4#r)  |  -32.87806    1.28575     -35.39808   -30.35803
      (1#3#r) vs (0#4#r)  |  -9.520418   .9335336     -11.35011   -7.690726
      (1#4#d) vs (0#4#r)  |  -2.903225   1.282248     -5.416384   -.3900657
      (1#4#o) vs (0#4#r)  |  -22.73753   .9971954     -24.69199   -20.78306
      (1#4#r) vs (0#4#r)  |  -1.347222   .9692299     -3.246877    .5524341
      (1#5#d) vs (0#4#r)  |  -2.701036   1.319182     -5.286585   -.1154876
      (1#5#o) vs (0#4#r)  |  -18.00009   .8689726     -19.70324   -16.29693
      (1#5#r) vs (0#4#r)  |   -2.17456   1.265695     -4.655276    .3061569
      (0#5#o) vs (0#5#d)  |  -13.17527   1.007915     -15.15075    -11.1998
      (0#5#r) vs (0#5#d)  |   2.466739   .8576915       .785694    4.147783
      (1#1#d) vs (0#5#d)  |  -28.08271   1.744408     -31.50169   -24.66374
      (1#1#o) vs (0#5#d)  |  -39.40249    1.66573     -42.66726   -36.13772
      (1#1#r) vs (0#5#d)  |  -23.99215   2.002205      -27.9164   -20.06791
      (1#2#d) vs (0#5#d)  |  -24.04143   1.547866      -27.0752   -21.00767
      (1#2#o) vs (0#5#d)  |  -39.28817   1.210209     -41.66014    -36.9162
      (1#2#r) vs (0#5#d)  |  -19.62933   1.474168     -22.51865   -16.74002
      (1#3#d) vs (0#5#d)  |  -11.76596   1.441005     -14.59027   -8.941638
      (1#3#o) vs (0#5#d)  |  -32.18048    1.04628     -34.23115   -30.12981
      (1#3#r) vs (0#5#d)  |  -8.822837   1.562063     -11.88443   -5.761249
      (1#4#d) vs (0#5#d)  |  -2.205644   1.316789     -4.786503    .3752155
      (1#4#o) vs (0#5#d)  |  -22.03995    .979275     -23.95929    -20.1206
      (1#4#r) vs (0#5#d)  |  -.6496404   1.548325     -3.684302    2.385021
      (1#5#d) vs (0#5#d)  |  -2.003455   .7772413      -3.52682     -.48009
      (1#5#o) vs (0#5#d)  |  -17.30251   .8209403     -18.91152   -15.69349
      (1#5#r) vs (0#5#d)  |  -1.476979   1.285423      -3.99636    1.042403
      (0#5#r) vs (0#5#o)  |   15.64201   .8589736      13.95845    17.32557
      (1#1#d) vs (0#5#o)  |  -14.90744    1.68993     -18.21964   -11.59524
      (1#1#o) vs (0#5#o)  |  -26.22722   1.524596     -29.21537   -23.23906
      (1#1#r) vs (0#5#o)  |  -10.81688   1.599055     -13.95097   -7.682791
      (1#2#d) vs (0#5#o)  |  -10.86616   1.868426     -14.52821   -7.204114
      (1#2#o) vs (0#5#o)  |   -26.1129   1.273171     -28.60827   -23.61753
      (1#2#r) vs (0#5#o)  |  -6.454062   1.291161     -8.984692   -3.923432
      (1#3#d) vs (0#5#o)  |   1.409317   1.753431     -2.027344    4.845978
      (1#3#o) vs (0#5#o)  |   -19.0052   1.136253     -21.23222   -16.77819
      (1#3#r) vs (0#5#o)  |   4.352436   1.436877      1.536209    7.168663
      (1#4#d) vs (0#5#o)  |   10.96963   1.771264      7.498015    14.44124
      (1#4#o) vs (0#5#o)  |  -8.864673    1.02415     -10.87197   -6.857376
      (1#4#r) vs (0#5#o)  |   12.52563   1.622969      9.344672    15.70659
      (1#5#d) vs (0#5#o)  |   11.17182   1.396403      8.434918    13.90872
      (1#5#o) vs (0#5#o)  |  -4.127233   .7659274     -5.628423   -2.626043
      (1#5#r) vs (0#5#o)  |   11.69829   1.340152      9.071644    14.32494
      (1#1#d) vs (0#5#r)  |  -30.54945   2.015075     -34.49893   -26.59998
      (1#1#o) vs (0#5#r)  |  -41.86923   1.842268     -45.48001   -38.25845
      (1#1#r) vs (0#5#r)  |  -26.45889   1.869338     -30.12273   -22.79506
      (1#2#d) vs (0#5#r)  |  -26.50817   1.997568     -30.42334   -22.59301
      (1#2#o) vs (0#5#r)  |  -41.75491    1.48879     -44.67288   -38.83693
      (1#2#r) vs (0#5#r)  |  -22.09607   1.423162     -24.88542   -19.30673
      (1#3#d) vs (0#5#r)  |  -14.23269   1.822848     -17.80541   -10.65998
      (1#3#o) vs (0#5#r)  |  -34.64722   1.471956      -37.5322   -31.76223
      (1#3#r) vs (0#5#r)  |  -11.28958   1.407241     -14.04772   -8.531435
      (1#4#d) vs (0#5#r)  |  -4.672382   1.683498     -7.971979   -1.372786
      (1#4#o) vs (0#5#r)  |  -24.50668   1.186552     -26.83228   -22.18109
      (1#4#r) vs (0#5#r)  |  -3.116379   1.354769     -5.771678   -.4610795
      (1#5#d) vs (0#5#r)  |  -4.470194   1.135702     -6.696129   -2.244258
      (1#5#o) vs (0#5#r)  |  -19.76924   .9003662     -21.53393   -18.00456
      (1#5#r) vs (0#5#r)  |  -3.943717   .9402797     -5.786631   -2.100803
      (1#1#o) vs (1#1#d)  |  -11.31977   1.469018       -14.199   -8.440552
      (1#1#r) vs (1#1#d)  |    4.09056   .9428141      2.242679    5.938442
      (1#2#d) vs (1#1#d)  |    4.04128   .9162628      2.245438    5.837122
      (1#2#o) vs (1#1#d)  |  -11.20546   1.422962     -13.99441   -8.416501
      (1#2#r) vs (1#1#d)  |    8.45338   .9667959      6.558495    10.34827
      (1#3#d) vs (1#1#d)  |   16.31676   1.226309      13.91324    18.72028
      (1#3#o) vs (1#1#d)  |  -4.097762   1.021417     -6.099703   -2.095822
      (1#3#r) vs (1#1#d)  |   19.25988   1.450901      16.41616    22.10359
      (1#4#d) vs (1#1#d)  |   25.87707   1.414462      23.10478    28.64937
      (1#4#o) vs (1#1#d)  |   6.042769   1.299236      3.496314    8.589225
      (1#4#r) vs (1#1#d)  |   27.43307   1.753664      23.99596    30.87019
      (1#5#d) vs (1#1#d)  |   26.07926    1.93624       22.2843    29.87422
      (1#5#o) vs (1#1#d)  |   10.78021   1.516241      7.808431    13.75199
      (1#5#r) vs (1#1#d)  |   26.60574   2.239522      22.21635    30.99512
      (1#1#r) vs (1#1#o)  |   15.41034   1.644638       12.1869    18.63377
      (1#2#d) vs (1#1#o)  |   15.36105   1.991968      11.45687    19.26524
      (1#2#o) vs (1#1#o)  |   .1143194    1.01222     -1.869595    2.098234
      (1#2#r) vs (1#1#o)  |   19.77316   1.797067      16.25097    23.29534
      (1#3#d) vs (1#1#o)  |   27.63653   2.235329      23.25537     32.0177
      (1#3#o) vs (1#1#o)  |   7.222012    1.15939      4.949649    9.494376
      (1#3#r) vs (1#1#o)  |   30.57965   2.240106      26.18913    34.97018
      (1#4#d) vs (1#1#o)  |   37.19685   2.307582      32.67407    41.71962
      (1#4#o) vs (1#1#o)  |   17.36254   1.559164      14.30664    20.41845
      (1#4#r) vs (1#1#o)  |   38.75285   2.460269      33.93081    43.57489
      (1#5#d) vs (1#1#o)  |   37.39903   2.224984      33.03815    41.75992
      (1#5#o) vs (1#1#o)  |   22.09998   1.430842      19.29558    24.90438
      (1#5#r) vs (1#1#o)  |   37.92551   2.444748      33.13389    42.71713
      (1#2#d) vs (1#1#r)  |  -.0492802   1.577796     -3.141704    3.043144
      (1#2#o) vs (1#1#r)  |  -15.29602   1.702656     -18.63316   -11.95887
      (1#2#r) vs (1#1#r)  |    4.36282   .8104045      2.774456    5.951184
      (1#3#d) vs (1#1#r)  |    12.2262   1.669922      8.953212    15.49919
      (1#3#o) vs (1#1#r)  |  -8.188323   1.465512     -11.06067   -5.315971
      (1#3#r) vs (1#1#r)  |   15.16932   1.230959      12.75668    17.58195
      (1#4#d) vs (1#1#r)  |   21.78651   1.794654      18.26905    25.30397
      (1#4#o) vs (1#1#r)  |   1.952209   1.529051     -1.044676    4.949094
      (1#4#r) vs (1#1#r)  |   23.34251   1.578937      20.24785    26.43717
      (1#5#d) vs (1#1#r)  |    21.9887   2.170281      17.73503    26.24237
      (1#5#o) vs (1#1#r)  |   6.689649   1.547656      3.656299    9.722998
      (1#5#r) vs (1#1#r)  |   22.51518    2.09966      18.39992    26.63043
      (1#2#o) vs (1#2#d)  |  -15.24674    1.68355     -18.54643   -11.94704
      (1#2#r) vs (1#2#d)  |     4.4121   1.144925      2.168088    6.656112
      (1#3#d) vs (1#2#d)  |   12.27548    .648214        11.005    13.54596
      (1#3#o) vs (1#2#d)  |  -8.139042   1.141611     -10.37656   -5.901527
      (1#3#r) vs (1#2#d)  |    15.2186   1.390631      12.49301    17.94418
      (1#4#d) vs (1#2#d)  |   21.83579   .8095179      20.24916    23.42242
      (1#4#o) vs (1#2#d)  |   2.001489   1.285539     -.5181199    4.521098
      (1#4#r) vs (1#2#d)  |   23.39179   1.527789      20.39738     26.3862
      (1#5#d) vs (1#2#d)  |   22.03798   1.547978        19.004    25.07196
      (1#5#o) vs (1#2#d)  |   6.738929   1.563971      3.673602    9.804256
      (1#5#r) vs (1#2#d)  |   22.56446   2.063724      18.51963    26.60928
      (1#2#r) vs (1#2#o)  |   19.65884   1.528336      16.66335    22.65432
      (1#3#d) vs (1#2#o)  |   27.52221   1.853557      23.88931    31.15512
      (1#3#o) vs (1#2#o)  |   7.107693    .842266      5.456882    8.758504
      (1#3#r) vs (1#2#o)  |   30.46533    1.89322      26.75469    34.17598
      (1#4#d) vs (1#2#o)  |   37.08253    1.88292      33.39207    40.77298
      (1#4#o) vs (1#2#o)  |   17.24822    1.20912      14.87839    19.61806
      (1#4#r) vs (1#2#o)  |   38.63853   2.081345      34.55917    42.71789
      (1#5#d) vs (1#2#o)  |   37.28472   1.617729      34.11402    40.45541
      (1#5#o) vs (1#2#o)  |   21.98566   1.068573       19.8913    24.08003
      (1#5#r) vs (1#2#o)  |   37.81119   1.960407      33.96886    41.65352
      (1#3#d) vs (1#2#r)  |   7.863379   1.022776      5.858775    9.867983
      (1#3#o) vs (1#2#r)  |  -12.55114   1.131659     -14.76915   -10.33313
      (1#3#r) vs (1#2#r)  |    10.8065    .538658      9.750747    11.86225
      (1#4#d) vs (1#2#r)  |   17.42369   1.112505      15.24322    19.60416
      (1#4#o) vs (1#2#r)  |  -2.410611   1.116834     -4.599565   -.2216567
      (1#4#r) vs (1#2#r)  |   18.97969   .8941245      17.22724    20.73215
      (1#5#d) vs (1#2#r)  |   17.62588   1.442721       14.7982    20.45356
      (1#5#o) vs (1#2#r)  |   2.326829   1.194501     -.0143507    4.668008
      (1#5#r) vs (1#2#r)  |   18.15236   1.434851       15.3401    20.96461
      (1#3#o) vs (1#3#d)  |  -20.41452   1.241304     -22.84743   -17.98161
      (1#3#r) vs (1#3#d)  |   2.943119   1.045797      .8933947    4.992843
      (1#4#d) vs (1#3#d)  |   9.560312   .4042252      8.768045    10.35258
      (1#4#o) vs (1#3#d)  |  -10.27399   1.221914      -12.6689   -7.879082
      (1#4#r) vs (1#3#d)  |   11.11632   1.140507      8.880962    13.35167
      (1#5#d) vs (1#3#d)  |   9.762501    1.23385      7.344198     12.1808
      (1#5#o) vs (1#3#d)  |   -5.53655   1.475818       -8.4291   -2.644001
      (1#5#r) vs (1#3#d)  |   10.28898   1.706451      6.944395    13.63356
      (1#3#r) vs (1#3#o)  |   23.35764   1.487833      20.44154    26.27374
      (1#4#d) vs (1#3#o)  |   29.97483   1.381174      27.26778    32.68188
      (1#4#o) vs (1#3#o)  |   10.14053   .8519977      8.470647    11.81042
      (1#4#r) vs (1#3#o)  |   31.53084   1.710346      28.17862    34.88305
      (1#5#d) vs (1#3#o)  |   30.17702   1.369775      27.49231    32.86173
      (1#5#o) vs (1#3#o)  |   14.87797   .8817111      13.14985    16.60609
      (1#5#r) vs (1#3#o)  |    30.7035   1.784461      27.20602    34.20098
      (1#4#d) vs (1#3#r)  |   6.617193   1.060523      4.538607     8.69578
      (1#4#o) vs (1#3#r)  |  -13.21711   1.342392     -15.84815   -10.58607
      (1#4#r) vs (1#3#r)  |   8.173197   .4743911      7.243407    9.102986
      (1#5#d) vs (1#3#r)  |   6.819382   1.332051      4.208609    9.430155
      (1#5#o) vs (1#3#r)  |  -8.479669   1.355824     -11.13703   -5.822303
      (1#5#r) vs (1#3#r)  |   7.345858    1.16992      5.052858    9.638859
      (1#4#o) vs (1#4#d)  |   -19.8343   1.246996     -22.27837   -17.39023
      (1#4#r) vs (1#4#d)  |   1.556003   1.007867     -.4193793    3.531386
      (1#5#d) vs (1#4#d)  |   .2021887   1.047297     -1.850476    2.254854
      (1#5#o) vs (1#4#d)  |  -15.09686   1.495294     -18.02758   -12.16614
      (1#5#r) vs (1#4#d)  |   .7286653   1.549793     -2.308873    3.766203
      (1#4#r) vs (1#4#o)  |    21.3903   1.458558      18.53158    24.24903
      (1#5#d) vs (1#4#o)  |   20.03649   1.301531      17.48554    22.58744
      (1#5#o) vs (1#4#o)  |    4.73744   .7989328       3.17156    6.303319
      (1#5#r) vs (1#4#o)  |   20.56297   1.606899       17.4135    23.71243
      (1#5#d) vs (1#4#r)  |  -1.353815     1.2267     -3.758103    1.050474
      (1#5#o) vs (1#4#r)  |  -16.65287   1.498695     -19.59025   -13.71548
      (1#5#r) vs (1#4#r)  |  -.8273381   1.024246     -2.834824    1.180148
      (1#5#o) vs (1#5#d)  |  -15.29905   1.253104     -17.75509   -12.84301
      (1#5#r) vs (1#5#d)  |   .5264765    .902638     -1.242661    2.295614
      (1#5#r) vs (1#5#o)  |   15.82553   1.462695       12.9587    18.69236
---------------------------------------------------------------------------

. 
. 
. 
. 
. 
. 
. 
. * Table 2: County-level analysis with control variables, estimating difference between turnout among older an
> d younger voters

. 
. 
. 
. * Import raw data for county-level analysis

. 
. import delimited "ky_county.csv", clear
(14 vars, 210 obs)

. 
. 
. 
. * Deskew population density by logarithmic transformation

. 
. gen lnpopdns = ln(popsqkm)

. 
. 
. 
. * Rescale household income so regression effect sizes are more readable and comparable to other variables'

. 
. replace hhinc2010 = hhinc2010/1000
(210 real changes made)

. 
. 
. 
. * Label variables

. 
. label var year "Year of election"

. 
. label var code "County, numbered alphabetically per Kentucky Secretary of State"

. 
. label var county "County name"

. 
. label var agegap1 "Percentage-point difference in turnout between under-25 and over-61 age groups"

. 
. label var agegap2 "Percentage-point difference in turnout between under-35 and over-49 age groups"

. 
. label var under25 "Percentage of population under age 25"

. 
. label var over65 "Percentage of population aged 65 or over"

. 
. label var totblk "Percentage of population that is Black (Hispanic or non-)"

. 
. label var tothisp "Percentage of population that is Hispanic (all races)"

. 
. label var unemp "County unemployment rate (source: Bureau of Labor Statistics)"

. 
. label var hhinc2010 "Median household income, deflated to 2010 dollars using Consumer Price Index"

. 
. label var popsqkm "Population density per square kilometer of land area"

. 
. label var timezone "Time Zone: 0 = Eastern Time; 1 = Central Time"

. 
. label var congdist "Congressional District: 1 = 1st; 2 = 2nd; 5 = 5th"

. 
. label var lnpopdns "Natural logarithm of population density: ln(popsqkm)"

. 
. 
. 
. * Set up panel by county (cross-section) and year (time-series)

. 
. xtset code year
       panel variable:  code (strongly balanced)
        time variable:  year, 1998 to 2016, but with gaps
                delta:  1 unit

. 
. 
. 
. * Estimate model for Table 2's Column I

. 
. xtpcse agegap1 timezone under25 over65 totblk tothisp lnpopdns unemp hhinc2010 congdist##year

Number of gaps in sample:  56

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   code                          Number of obs     =        196
Time variable:    year                          Number of groups  =         14
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =         14
                                                              avg =         14
                                                              max =         14
Estimated covariances      =       105          R-squared         =     0.9219
Estimated autocorrelations =         0          Wald chi2(22)     =    1423.21
Estimated coefficients     =        50          Prob > chi2       =     0.0000

-------------------------------------------------------------------------------
              |           Panel-corrected
      agegap1 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |  -1.449946   .5691333    -2.55   0.011    -2.565427   -.3344653
      under25 |   1.160028   .5745716     2.02   0.043     .0338884    2.286168
       over65 |   .6143054   .4940897     1.24   0.214    -.3540927    1.582704
       totblk |   .0638612   .1345016     0.47   0.635     -.199757    .3274795
      tothisp |  -.5984408   .4420541    -1.35   0.176    -1.464851    .2679693
     lnpopdns |   1.047812   1.121126     0.93   0.350    -1.149555    3.245179
        unemp |   .0322809   .1163958     0.28   0.782    -.1958508    .2604125
    hhinc2010 |   .7638649   .0867828     8.80   0.000     .5937739     .933956
              |
     congdist |
           2  |   1.339785   1.410649     0.95   0.342    -1.425036    4.104605
           5  |   -3.87631   1.244879    -3.11   0.002    -6.316228   -1.436392
              |
         year |
        1999  |  -8.835853   1.420679    -6.22   0.000    -11.62033   -6.051373
        2000  |   -5.76349   1.838023    -3.14   0.002    -9.365949   -2.161031
        2003  |   4.614385   1.543987     2.99   0.003     1.588226    7.640544
        2004  |  -12.78467   1.650309    -7.75   0.000    -16.01921    -9.55012
        2006  |   6.774863   1.698947     3.99   0.000     3.444988    10.10474
        2007  |   4.552617   1.713307     2.66   0.008     1.194598    7.910636
        2008  |  -7.102857    1.67701    -4.24   0.000    -10.38974   -3.815977
        2010  |    10.4153   1.395658     7.46   0.000     7.679864    13.15074
        2011  |   9.175492   1.403956     6.54   0.000     6.423789     11.9272
        2012  |   1.229574   1.476829     0.83   0.405    -1.664959    4.124106
        2014  |   10.94907   1.736143     6.31   0.000     7.546291    14.35184
        2015  |   5.531387   1.849548     2.99   0.003     1.906338    9.156435
        2016  |  -10.33422   2.024315    -5.11   0.000    -14.30181   -6.366638
              |
congdist#year |
      2 1999  |  -.0535828   1.572281    -0.03   0.973    -3.135197    3.028031
      2 2000  |  -5.130734   1.660996    -3.09   0.002    -8.386227   -1.875242
      2 2003  |  -1.611848   1.510755    -1.07   0.286    -4.572873    1.349177
      2 2004  |   -2.85271   1.569266    -1.82   0.069    -5.928415    .2229944
      2 2006  |   -2.16681   1.636948    -1.32   0.186     -5.37517    1.041549
      2 2007  |  -2.219282   1.656479    -1.34   0.180    -5.465921    1.027357
      2 2008  |  -6.497528   1.715287    -3.79   0.000    -9.859429   -3.135626
      2 2010  |  -.5052801   1.752018    -0.29   0.773    -3.939172    2.928612
      2 2011  |  -3.736061   1.708742    -2.19   0.029    -7.085134   -.3869885
      2 2012  |  -3.888215   1.713397    -2.27   0.023     -7.24641   -.5300188
      2 2014  |  -1.350632   1.778399    -0.76   0.448     -4.83623    2.134966
      2 2015  |  -1.173517   1.765283    -0.66   0.506    -4.633408    2.286374
      2 2016  |   .1656543   1.825667     0.09   0.928    -3.412587    3.743896
      5 1999  |   3.841413   1.175559     3.27   0.001     1.537359    6.145466
      5 2000  |   4.044758   1.434116     2.82   0.005     1.233943    6.855572
      5 2003  |    4.77794   1.269283     3.76   0.000     2.290191    7.265689
      5 2004  |    4.00756   1.287284     3.11   0.002      1.48453    6.530591
      5 2006  |   3.079611   1.390318     2.22   0.027     .3546388    5.804584
      5 2007  |   4.262242   1.389749     3.07   0.002     1.538383    6.986101
      5 2008  |    6.67227    1.42445     4.68   0.000     3.880399    9.464141
      5 2010  |   5.234049   1.457886     3.59   0.000     2.376645    8.091454
      5 2011  |   1.898076   1.509955     1.26   0.209    -1.061381    4.857533
      5 2012  |   5.934948   1.464806     4.05   0.000     3.063981    8.805915
      5 2014  |   5.428786   1.390165     3.91   0.000     2.704113    8.153459
      5 2015  |   2.767024   1.353535     2.04   0.041     .1141433    5.419904
      5 2016  |   4.660429   1.430272     3.26   0.001     1.857148    7.463711
              |
        _cons |  -41.50841   24.19337    -1.72   0.086    -88.92654    5.909714
-------------------------------------------------------------------------------

. 
. 
. 
. * Estimate model for Table 2's Column II

. 
. xtpcse agegap2 timezone under25 over65 totblk tothisp lnpopdns unemp hhinc2010 congdist##year

Number of gaps in sample:  56

Linear regression, correlated panels corrected standard errors (PCSEs)

Group variable:   code                          Number of obs     =        196
Time variable:    year                          Number of groups  =         14
Panels:           correlated (balanced)         Obs per group:
Autocorrelation:  no autocorrelation                          min =         14
                                                              avg =         14
                                                              max =         14
Estimated covariances      =       105          R-squared         =     0.9057
Estimated autocorrelations =         0          Wald chi2(22)     =     955.29
Estimated coefficients     =        50          Prob > chi2       =     0.0000

-------------------------------------------------------------------------------
              |           Panel-corrected
      agegap2 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
--------------+----------------------------------------------------------------
     timezone |  -.9795711   .4013972    -2.44   0.015    -1.766295    -.192847
      under25 |   .9602259   .4687322     2.05   0.041     .0415276    1.878924
       over65 |    .432426   .3934351     1.10   0.272    -.3386927    1.203545
       totblk |    .065911   .0974354     0.68   0.499     -.125059     .256881
      tothisp |  -.2416925   .2991927    -0.81   0.419    -.8280994    .3447144
     lnpopdns |   1.468283   .6704202     2.19   0.029     .1542836    2.782282
        unemp |  -.0393619   .1167583    -0.34   0.736     -.268204    .1894802
    hhinc2010 |   .4869275   .0639714     7.61   0.000     .3615458    .6123092
              |
     congdist |
           2  |   1.346375   1.075745     1.25   0.211    -.7620475    3.454797
           5  |  -3.058245    1.04098    -2.94   0.003    -5.098528   -1.017963
              |
         year |
        1999  |  -9.809155   1.201311    -8.17   0.000    -12.16368   -7.454627
        2000  |  -4.664496   1.619775    -2.88   0.004    -7.839197   -1.489795
        2003  |   3.199032   1.383804     2.31   0.021     .4868264    5.911238
        2004  |   -8.00341   1.488517    -5.38   0.000    -10.92085   -5.085971
        2006  |   3.741261   1.519612     2.46   0.014     .7628759    6.719645
        2007  |   1.909235   1.536683     1.24   0.214    -1.102609    4.921079
        2008  |   -4.21326   1.457136    -2.89   0.004    -7.069195   -1.357325
        2010  |   7.459312   1.072181     6.96   0.000     5.357877    9.560748
        2011  |   4.665874   1.099388     4.24   0.000     2.511114    6.820635
        2012  |    1.30034   1.189332     1.09   0.274    -1.030708    3.631387
        2014  |     7.9303   1.368072     5.80   0.000     5.248928    10.61167
        2015  |   2.965142   1.442895     2.05   0.040     .1371205    5.793164
        2016  |  -5.612137   1.662475    -3.38   0.001    -8.870528   -2.353747
              |
congdist#year |
      2 1999  |   2.209042   1.231899     1.79   0.073    -.2054367     4.62352
      2 2000  |  -2.094158   1.297612    -1.61   0.107    -4.637431    .4491154
      2 2003  |  -1.025035   1.247774    -0.82   0.411    -3.470627    1.420557
      2 2004  |  -2.514544   1.283182    -1.96   0.050    -5.029535    .0004465
      2 2006  |  -1.184194   1.328749    -0.89   0.373    -3.788496    1.420107
      2 2007  |   -1.03774   1.348089    -0.77   0.441    -3.679947    1.604467
      2 2008  |  -5.143165   1.399219    -3.68   0.000    -7.885583   -2.400747
      2 2010  |  -1.115472   1.397788    -0.80   0.425    -3.855086    1.624141
      2 2011  |  -2.288678   1.343145    -1.70   0.088    -4.921194    .3438372
      2 2012  |  -3.336257   1.375544    -2.43   0.015    -6.032274   -.6402398
      2 2014  |  -2.206035   1.389575    -1.59   0.112    -4.929552    .5174821
      2 2015  |   -2.03762   1.364359    -1.49   0.135    -4.711714    .6364741
      2 2016  |  -1.363967   1.426569    -0.96   0.339     -4.15999    1.432056
      5 1999  |    3.58073     .91984     3.89   0.000     1.777877    5.383583
      5 2000  |   3.307704   1.179726     2.80   0.005     .9954848    5.619924
      5 2003  |   2.104695   1.083619     1.94   0.052      -.01916     4.22855
      5 2004  |   1.875854   1.082526     1.73   0.083    -.2458574    3.997566
      5 2006  |    1.90195   1.182741     1.61   0.108    -.4161805     4.22008
      5 2007  |   3.207365   1.179895     2.72   0.007      .894813    5.519916
      5 2008  |   2.825395   1.206847     2.34   0.019      .460017    5.190772
      5 2010  |   2.761242   1.206552     2.29   0.022     .3964427     5.12604
      5 2011  |   1.433654   1.260843     1.14   0.256    -1.037552     3.90486
      5 2012  |   3.143421   1.234718     2.55   0.011     .7234181    5.563424
      5 2014  |   2.770394   1.135257     2.44   0.015     .5453303    4.995458
      5 2015  |   1.052373   1.074658     0.98   0.327    -1.053918    3.158664
      5 2016  |   2.712587   1.156773     2.34   0.019     .4453536    4.979821
              |
        _cons |  -28.46951   20.18288    -1.41   0.158    -68.02723     11.0882
-------------------------------------------------------------------------------

. 
. 
. 
. 
. 
. 
. 
. * Appendix: Falsification analysis

. 
. 
. 
. xtreg under25 timezone, be

Between regression (regression on group means)  Number of obs     =        210
Group variable: code                            Number of groups  =         14

R-sq:                                           Obs per group:
     within  = 0.0000                                         min =         15
     between = 0.1124                                         avg =       15.0
     overall = 0.0748                                         max =         15

                                                F(1,12)           =       1.52
sd(u_i + avg(e_i.))=  1.533568                  Prob > F          =     0.2414

------------------------------------------------------------------------------
     under25 |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
    timezone |  -1.010259   .8197267    -1.23   0.241     -2.79629    .7757722
       _cons |   26.97098   .5796343    46.53   0.000     25.70807     28.2339
------------------------------------------------------------------------------

. 
. xtreg over65 timezone, be

Between regression (regression on group means)  Number of obs     =        210
Group variable: code                            Number of groups  =         14

R-sq:                                           Obs per group:
     within  = 0.0000                                         min =         15
     between = 0.1466                                         avg =       15.0
     overall = 0.1096                                         max =         15

                                                F(1,12)           =       2.06
sd(u_i + avg(e_i.))=  2.105394                  Prob > F          =     0.1767

------------------------------------------------------------------------------
      over65 |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
    timezone |   1.615626    1.12538     1.44   0.177    -.8363675    4.067619
       _cons |   14.35858   .7957641    18.04   0.000     12.62476     16.0924
------------------------------------------------------------------------------

. 
. xtreg totblk timezone, be

Between regression (regression on group means)  Number of obs     =        210
Group variable: code                            Number of groups  =         14

R-sq:                                           Obs per group:
     within  = 0.0000                                         min =         15
     between = 0.1055                                         avg =       15.0
     overall = 0.1030                                         max =         15

                                                F(1,12)           =       1.42
sd(u_i + avg(e_i.))=  3.200842                  Prob > F          =     0.2571

------------------------------------------------------------------------------
      totblk |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
    timezone |  -2.035668   1.710922    -1.19   0.257    -5.763447    1.692112
       _cons |   4.356597   1.209805     3.60   0.004     1.720658    6.992535
------------------------------------------------------------------------------

. 
. xtreg tothisp timezone, be

Between regression (regression on group means)  Number of obs     =        210
Group variable: code                            Number of groups  =         14

R-sq:                                           Obs per group:
     within  = 0.0000                                         min =         15
     between = 0.2987                                         avg =       15.0
     overall = 0.1995                                         max =         15

                                                F(1,12)           =       5.11
sd(u_i + avg(e_i.))=  .8308321                  Prob > F          =     0.0432

------------------------------------------------------------------------------
     tothisp |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
    timezone |  -1.003978   .4440984    -2.26   0.043    -1.971586   -.0363711
       _cons |   2.411733    .314025     7.68   0.000     1.727531    3.095935
------------------------------------------------------------------------------

. 
. xtreg lnpopdns timezone, be

Between regression (regression on group means)  Number of obs     =        210
Group variable: code                            Number of groups  =         14

R-sq:                                           Obs per group:
     within  = 0.0000                                         min =         15
     between = 0.2662                                         avg =       15.0
     overall = 0.2644                                         max =         15

                                                F(1,12)           =       4.35
sd(u_i + avg(e_i.))=  .3993875                  Prob > F          =     0.0589

------------------------------------------------------------------------------
    lnpopdns |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
    timezone |  -.4454158   .2134816    -2.09   0.059    -.9105523    .0197207
       _cons |   3.323636   .1509543    22.02   0.000     2.994735    3.652538
------------------------------------------------------------------------------

. 
. xtreg unemp timezone, be

Between regression (regression on group means)  Number of obs     =        210
Group variable: code                            Number of groups  =         14

R-sq:                                           Obs per group:
     within  =      .                                         min =         15
     between = 0.1391                                         avg =       15.0
     overall = 0.0164                                         max =         15

                                                F(1,12)           =       1.94
sd(u_i + avg(e_i.))=  1.081022                  Prob > F          =     0.1890

------------------------------------------------------------------------------
       unemp |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
    timezone |   .8047619   .5778308     1.39   0.189    -.4542233    2.063747
       _cons |   7.089524   .4085881    17.35   0.000     6.199287    7.979761
------------------------------------------------------------------------------

. 
. xtreg hhinc2010 timezone, be

Between regression (regression on group means)  Number of obs     =        210
Group variable: code                            Number of groups  =         14

R-sq:                                           Obs per group:
     within  = 0.0000                                         min =         15
     between = 0.1736                                         avg =       15.0
     overall = 0.1652                                         max =         15

                                                F(1,12)           =       2.52
sd(u_i + avg(e_i.))=  6.376331                  Prob > F          =     0.1384

------------------------------------------------------------------------------
   hhinc2010 |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
    timezone |  -5.410713   3.408292    -1.59   0.138    -12.83674    2.015318
       _cons |   37.28996   2.410027    15.47   0.000     32.03897    42.54096
------------------------------------------------------------------------------

. log close
---------------------------------------------------------------------------------------------------------------
